- updated to 2.17.50.0.2.

[packages/binutils.git] / binutils-gasp.patch

diff -urNbB binutils-2.16.91.0.2.org/gas/Makefile.am binutils-2.16.91.0.2/gas/Makefile.am
--- binutils-2.16.91.0.2.org/gas/Makefile.am	2005-07-20 19:27:27.000000000 +0000
+++ binutils-2.16.91.0.2/gas/Makefile.am	2005-07-21 18:35:02.000000000 +0000
@@ -204,7 +204,7 @@
 	symbols.c \
 	write.c
 
-CFILES = $(GAS_CFILES) itbl-ops.c
+CFILES = $(GAS_CFILES) gasp.c itbl-ops.c
 
 HFILES = \
 	as.h \
@@ -466,7 +466,8 @@
 
 # Note: GASP is now deprecated and has been removed.  It is still
 # available in the CVS archive or older binutils releases if it is needed.
-noinst_PROGRAMS = as-new
+# ...and it is needed for few packages in distribution.
+noinst_PROGRAMS = as-new gasp-new
 noinst_SCRIPTS = $(GDBINIT)
 EXTRA_SCRIPTS = .gdbinit
 
@@ -530,6 +531,10 @@
 	as.h asintl.h bignum.h bit_fix.h config.h emul.h expr.h flonum.h \
 	frags.h hash.h listing.h obj.h read.h symbols.h tc.h write.h
 
+gasp_new_SOURCES = gasp.c macro.c sb.c hash.c
+gasp_new_LDADD = ../libiberty/libiberty.a $(INTLLIBS)
+gasp_new_DEPENDENCIES = ../libiberty/libiberty.a $(INTLDEPS)
+
 EXPECT = expect
 RUNTEST = runtest
 RUNTESTFLAGS=
@@ -2508,6 +2513,8 @@
   $(INCDIR)/obstack.h subsegs.h struc-symbol.h
 write.o: write.c $(INCDIR)/symcat.h subsegs.h $(INCDIR)/obstack.h \
   output-file.h dwarf2dbg.h
+gasp.o: gasp.c $(INCDIR)/getopt.h $(INCDIR)/safe-ctype.h \
+  sb.h macro.h $(INCDIR)/xregex.h $(INCDIR)/xregex2.h
 itbl-ops.o: itbl-ops.c itbl-ops.h $(INCDIR)/symcat.h
 e-crisaout.o: $(srcdir)/config/e-crisaout.c $(INCDIR)/symcat.h \
   emul-target.h
diff -urNbB binutils-2.16.91.0.2.org/gas/doc/Makefile.am binutils-2.16.91.0.2/gas/doc/Makefile.am
--- binutils-2.16.91.0.2.org/gas/doc/Makefile.am	2005-07-20 19:27:27.000000000 +0000
+++ binutils-2.16.91.0.2/gas/doc/Makefile.am	2005-07-21 18:31:04.000000000 +0000
@@ -15,7 +15,7 @@
 
 man_MANS = as.1
 
-info_TEXINFOS = as.texinfo 
+info_TEXINFOS = as.texinfo gasp.texinfo
 
 asconfig.texi: $(CONFIG).texi
 	rm -f asconfig.texi
diff -urNbB binutils-2.16.91.0.2.org/gas/doc/gasp.texinfo binutils-2.16.91.0.2/gas/doc/gasp.texinfo
--- binutils-2.16.91.0.2.org/gas/doc/gasp.texinfo	1970-01-01 00:00:00.000000000 +0000
+++ binutils-2.16.91.0.2/gas/doc/gasp.texinfo	2005-07-21 18:31:04.000000000 +0000
@@ -0,0 +1,1456 @@
+\input texinfo               @c             -*- Texinfo -*-
+@setfilename gasp.info
+@c
+@c This file documents the assembly preprocessor "GASP"
+@c
+@c Copyright 1994, 1995, 2000, 2002 Free Software Foundation, Inc.
+@c
+@c    Permission is granted to copy, distribute and/or modify this document
+@c    under the terms of the GNU Free Documentation License, Version 1.1
+@c    or any later version published by the Free Software Foundation;
+@c    with no Invariant Sections, with no Front-Cover Texts, and with no
+@c    Back-Cover Texts.  A copy of the license is included in the
+@c    section entitled "GNU Free Documentation License".
+
+@ifinfo
+@format
+START-INFO-DIR-ENTRY
+* gasp: (gasp).                     The GNU Assembler Preprocessor
+END-INFO-DIR-ENTRY
+@end format
+@end ifinfo
+
+@syncodeindex ky cp
+@syncodeindex fn cp
+
+@finalout
+@setchapternewpage odd
+@settitle GASP
+@titlepage
+@c FIXME boring title
+@title GASP, an assembly preprocessor
+@subtitle for GASP version 1
+@sp 1
+@subtitle March 1994
+@author Roland Pesch
+@page
+
+@tex
+{\parskip=0pt \hfill Cygnus Support\par
+}
+@end tex
+
+@vskip 0pt plus 1filll
+Copyright @copyright{} 1994, 1995, 2000, 2002 Free Software Foundation, Inc.
+
+      Permission is granted to copy, distribute and/or modify this document
+      under the terms of the GNU Free Documentation License, Version 1.1
+      or any later version published by the Free Software Foundation;
+      with no Invariant Sections, with no Front-Cover Texts, and with no
+      Back-Cover Texts.  A copy of the license is included in the
+      section entitled "GNU Free Documentation License".
+
+@end titlepage
+
+@ifinfo
+Copyright @copyright{} 1994, 1995, 2000, 2002 Free Software Foundation, Inc.
+
+@ignore
+Permission is granted to process this file through TeX and print the
+results, provided the printed document carries a copying permission
+notice identical to this one except for the removal of this paragraph
+(this paragraph not being relevant to the printed manual).
+@end ignore
+
+      Permission is granted to copy, distribute and/or modify this document
+      under the terms of the GNU Free Documentation License, Version 1.1
+      or any later version published by the Free Software Foundation;
+      with no Invariant Sections, with no Front-Cover Texts, and with no
+      Back-Cover Texts.  A copy of the license is included in the
+      section entitled "GNU Free Documentation License".
+
+
+@node Top
+@top GASP
+
+GASP is a preprocessor for assembly programs.
+
+This file describes version 1 of GASP.
+
+Steve Chamberlain wrote GASP; Roland Pesch wrote this manual.
+
+@menu
+* Overview::                    What is GASP?
+* Invoking GASP::               Command line options.
+* Commands::                    Preprocessor commands.
+* GNU Free Documentation License::  GNU Free Documentation License
+* Index::                       Index.
+@end menu
+@end ifinfo
+
+@node Overview
+@chapter What is GASP?
+
+The primary purpose of the @sc{gnu} assembler is to assemble the output of
+other programs---notably compilers.  When you have to hand-code
+specialized routines in assembly, that means the @sc{gnu} assembler is
+an unfriendly processor: it has no directives for macros, conditionals,
+or many other conveniences that you might expect.
+
+In some cases you can simply use the C preprocessor, or a generalized
+preprocessor like @sc{m4}; but this can be awkward, since none of these
+things are designed with assembly in mind.
+
+@sc{gasp} fills this need.  It is expressly designed to provide the
+facilities you need with hand-coded assembly code.  Implementing it as a
+preprocessor, rather than part of the assembler, allows the maximum
+flexibility: you can use it with hand-coded assembly, without paying a
+penalty of added complexity in the assembler you use for compiler
+output.
+
+@emph{Note} The use of @sc{gasp} has now been deprecated.  Anything
+that it could do can now be done by the macro facilities built into
+@sc{gas} itself.  At some point in the future the @sc{gasp} sources will
+be removed entirely from the binutils distribution.
+
+Here is a small example to give the flavor of @sc{gasp}.  This input to
+@sc{gasp}
+
+@cartouche
+@example
+        .MACRO  saveregs from=8 to=14
+count   .ASSIGNA \from
+        ! save r\from..r\to
+        .AWHILE  \&count LE \to
+        mov     r\&count,@@-sp
+count   .ASSIGNA  \&count + 1
+        .AENDW
+        .ENDM
+
+        saveregs from=12
+
+bar:    mov     #H'dead+10,r0
+foo     .SDATAC "hello"<10>
+        .END
+@end example
+@end cartouche
+
+@noindent
+generates this assembly program:
+
+@cartouche
+@example
+        ! save r12..r14
+        mov     r12,@@-sp
+        mov     r13,@@-sp
+        mov     r14,@@-sp
+
+bar:    mov     #57005+10,r0
+foo:    .byte   6,104,101,108,108,111,10
+@end example
+@end cartouche
+
+@node Invoking GASP
+@chapter Command Line Options
+
+@c FIXME!  Or is there a simpler way, calling from GAS option?
+The simplest way to use @sc{gasp} is to run it as a filter and assemble
+its output.  In Unix and its ilk, you can do this, for example:
+
+@c FIXME! GASP filename suffix convention?
+@example
+$ gasp prog.asm | as -o prog.o
+@end example
+
+Naturally, there are also a few command-line options to allow you to
+request variations on this basic theme.  Here is the full set of
+possibilities for the @sc{gasp} command line.
+
+@example
+gasp  [ -a | --alternate ]
+      [ -c @var{char} | --commentchar @var{char} ]
+      [ -d | --debug ]  [ -h | --help ] [ -M | --mri ]
+      [ -o @var{outfile} | --output @var{outfile} ]
+      [ -p | --print ]  [ -s | --copysource ]
+      [ -u | --unreasonable ]  [ -v | --version ]
+      @var{infile} @dots{}
+@end example
+
+@ftable @code
+@item @var{infile} @dots{}
+@c FIXME! Why not stdin as default infile?
+The input file names.  You must specify at least one input file; if you
+specify more, @sc{gasp} preprocesses them all, concatenating the output
+in the order you list the @var{infile} arguments.
+
+Mark the end of each input file with the preprocessor command
+@code{.END}.  @xref{Other Commands,, Miscellaneous commands}.
+
+@item -a
+@itemx --alternate
+Use alternative macro syntax.  @xref{Alternate,, Alternate macro
+syntax}, for a discussion of how this syntax differs from the default
+@sc{gasp} syntax.
+
+@cindex comment character, changing
+@cindex semicolon, as comment
+@cindex exclamation mark, as comment
+@cindex shriek, as comment
+@cindex bang, as comment
+@cindex @code{!} default comment char
+@cindex @code{;} as comment char
+@item -c '@var{char}'
+@itemx --commentchar '@var{char}'
+Use @var{char} as the comment character.  The default comment character
+is @samp{!}.  For example, to use a semicolon as the comment character,
+specify @w{@samp{-c ';'}} on the @sc{gasp} command line.  Since
+assembler command characters often have special significance to command
+shells, it is a good idea to quote or escape @var{char} when you specify
+a comment character.
+
+For the sake of simplicity, all examples in this manual use the default
+comment character @samp{!}.
+
+@item -d
+@itemx --debug
+Show debugging statistics.  In this version of @sc{gasp}, this option
+produces statistics about the string buffers that @sc{gasp} allocates
+internally.  For each defined buffersize @var{s}, @sc{gasp} shows the
+number of strings @var{n} that it allocated, with a line like this:
+
+@example
+strings size @var{s} : @var{n}
+@end example
+
+@noindent
+@sc{gasp} displays these statistics on the standard error stream, when
+done preprocessing.
+
+@item -h
+@itemx --help
+Display a summary of the @sc{gasp} command line options.
+
+@item -M
+@itemx --mri
+Use MRI compatibility mode.  Using this option causes @sc{gasp} to
+accept the syntax and pseudo-ops used by the Microtec Research
+@code{ASM68K} assembler.
+
+@item -o @var{outfile}
+@itemx --output @var{outfile}
+Write the output in a file called @var{outfile}.  If you do not use the
+@samp{-o} option, @sc{gasp} writes its output on the standard output
+stream.
+
+@item -p
+@itemx --print
+Print line numbers.  @sc{gasp} obeys this option @emph{only} if you also
+specify @samp{-s} to copy source lines to its output.  With @samp{-s
+-p}, @sc{gasp} displays the line number of each source line copied
+(immediately after the comment character at the beginning of the line).
+
+@item -s
+@itemx --copysource
+Copy the source lines to the output file.  Use this option
+to see the effect of each preprocessor line on the @sc{gasp} output.
+@sc{gasp} places a comment character (@samp{!} by default) at
+the beginning of each source line it copies, so that you can use this
+option and still assemble the result.
+
+@item -u
+@itemx --unreasonable
+Bypass ``unreasonable expansion'' limit.  Since you can define @sc{gasp}
+macros inside other macro definitions, the preprocessor normally
+includes a sanity check.  If your program requires more than 1,000
+nested expansions, @sc{gasp} normally exits with an error message.  Use
+this option to turn off this check, allowing unlimited nested
+expansions.
+
+@item -v
+@itemx --version
+Display the @sc{gasp} version number.
+@end ftable
+
+@node Commands
+@chapter Preprocessor Commands
+
+@sc{gasp} commands have a straightforward syntax that fits in well with
+assembly conventions.  In general, a command extends for a line, and may
+have up to three fields: an optional label, the command itself, and
+optional arguments to the command.  You can write commands in upper or
+lower case, though this manual shows them in upper case.  @xref{Syntax
+Details,, Details of the GASP syntax}, for more information.
+
+@menu
+* Conditionals::
+* Loops::
+* Variables::
+* Macros::
+* Data::
+* Listings::
+* Other Commands::
+* Syntax Details::
+* Alternate::
+@end menu
+
+@node Conditionals
+@section Conditional assembly
+
+The conditional-assembly directives allow you to include or exclude
+portions of an assembly depending on how a pair of expressions, or a
+pair of strings, compare.
+
+The overall structure of conditionals is familiar from many other
+contexts.  @code{.AIF} marks the start of a conditional, and precedes
+assembly for the case when the condition is true.   An optional
+@code{.AELSE} precedes assembly for the converse case, and an
+@code{.AENDI} marks the end of the condition.
+
+@c FIXME! Why doesn't -u turn off this check?
+You may nest conditionals up to a depth of 100; @sc{gasp} rejects
+nesting beyond that, because it may indicate a bug in your macro
+structure.
+
+@c FIXME! Why isn't there something like cpp's -D option?  Conditionals
+@c        would be much more useful if there were.
+Conditionals are primarily useful inside macro definitions, where you
+often need different effects depending on argument values.
+@xref{Macros,, Defining your own directives}, for details about defining
+macros.
+
+@ftable @code
+@item .AIF @var{expra} @var{cmp} @var{exprb}
+@itemx .AIF "@var{stra}" @var{cmp} "@var{strb}"
+
+The governing condition goes on the same line as the @code{.AIF}
+preprocessor command.  You may compare either two strings, or two
+expressions.
+
+When you compare strings, only two conditional @var{cmp} comparison
+operators are available: @samp{EQ} (true if @var{stra} and @var{strb}
+are identical), and @samp{NE} (the opposite).
+
+When you compare two expressions, @emph{both expressions must be
+absolute} (@pxref{Expressions,, Arithmetic expressions in GASP}).  You
+can use these @var{cmp} comparison operators with expressions:
+
+@ftable @code
+@item EQ
+Are @var{expra} and @var{exprb} equal?  (For strings, are @var{stra} and
+@var{strb} identical?)
+
+@item NE
+Are @var{expra} and @var{exprb} different?  (For strings, are @var{stra}
+and @var{strb} different?
+
+@item LT
+Is @var{expra} less than @var{exprb}?  (Not allowed for strings.)
+
+@item LE
+Is @var{expra} less than or equal to @var{exprb}?  (Not allowed for strings.)
+
+@item GT
+Is @var{expra} greater than @var{exprb}?  (Not allowed for strings.)
+
+@item GE
+Is @var{expra} greater than or equal to @var{exprb}?  (Not allowed for
+strings.)
+@end ftable
+
+@item .AELSE
+Marks the start of assembly code to be included if the condition fails.
+Optional, and only allowed within a conditional (between @code{.AIF} and
+@code{.AENDI}).
+
+@item .AENDI
+Marks the end of a conditional assembly.
+@end ftable
+
+@node Loops
+@section Repetitive sections of assembly
+
+Two preprocessor directives allow you to repeatedly issue copies of the
+same block of assembly code.
+
+@ftable @code
+@item .AREPEAT @var{aexp}
+@itemx .AENDR
+If you simply need to repeat the same block of assembly over and over a
+fixed number of times, sandwich one instance of the repeated block
+between @code{.AREPEAT} and @code{.AENDR}.  Specify the number of
+copies as @var{aexp} (which must be an absolute expression).  For
+example, this repeats two assembly statements three times in succession:
+
+@cartouche
+@example
+        .AREPEAT        3
+        rotcl   r2
+        div1    r0,r1
+        .AENDR
+@end example
+@end cartouche
+
+@item .AWHILE @var{expra} @var{cmp} @var{exprb}
+@itemx .AENDW
+@itemx .AWHILE @var{stra} @var{cmp} @var{strb}
+@itemx .AENDW
+To repeat a block of assembly depending on a conditional test, rather
+than repeating it for a specific number of times, use @code{.AWHILE}.
+@code{.AENDW} marks the end of the repeated block.  The conditional
+comparison works exactly the same way as for @code{.AIF}, with the same
+comparison operators (@pxref{Conditionals,, Conditional assembly}).
+
+Since the terms of the comparison must be absolute expression,
+@code{.AWHILE} is primarily useful within macros.  @xref{Macros,,
+Defining your own directives}.
+@end ftable
+
+@cindex loops, breaking out of
+@cindex breaking out of loops
+You can use the @code{.EXITM} preprocessor directive to break out of
+loops early (as well as to break out of macros).  @xref{Macros,,
+Defining your own directives}.
+
+@node Variables
+@section Preprocessor variables
+
+You can use variables in @sc{gasp} to represent strings, registers, or
+the results of expressions.
+
+You must distinguish two kinds of variables: 
+@enumerate
+@item
+Variables defined with @code{.EQU} or @code{.ASSIGN}.  To evaluate this
+kind of variable in your assembly output, simply mention its name.  For
+example, these two lines define and use a variable @samp{eg}:
+
+@cartouche
+@example
+eg     .EQU   FLIP-64
+       @dots{}
+       mov.l  eg,r0
+@end example
+@end cartouche
+
+@emph{Do not use} this kind of variable in conditional expressions or
+while loops; @sc{gasp} only evaluates these variables when writing
+assembly output.
+
+@item
+Variables for use during preprocessing.  You can define these
+with @code{.ASSIGNC} or @code{.ASSIGNA}.  To evaluate this
+kind of variable, write @samp{\&} before the variable name; for example,
+
+@cartouche
+@example
+opcit  .ASSIGNA  47
+       @dots{}
+       .AWHILE  \&opcit GT 0
+       @dots{}
+       .AENDW
+@end example
+@end cartouche
+
+@sc{gasp} treats macro arguments almost the same way, but to evaluate
+them you use the prefix @samp{\} rather than @samp{\&}.
+@xref{Macros,, Defining your own directives}.
+@end enumerate
+
+@ftable @code
+@item @var{pvar} .EQU @var{expr}
+@c FIXME!  Anything to beware of re GAS directive of same name?
+Assign preprocessor variable @var{pvar} the value of the expression
+@var{expr}.  There are no restrictions on redefinition; use @samp{.EQU}
+with the same @var{pvar} as often as you find it convenient.
+
+@item @var{pvar} .ASSIGN @var{expr}
+Almost the same as @code{.EQU}, save that you may not redefine
+@var{pvar} using @code{.ASSIGN} once it has a value.
+@c FIXME!!  Supposed to work this way, apparently, but on 9feb94 works
+@c          just like .EQU
+
+@item @var{pvar} .ASSIGNA @var{aexpr}
+Define a variable with a numeric value, for use during preprocessing.
+@var{aexpr} must be an absolute expression.  You can redefine variables
+with @code{.ASSIGNA} at any time.
+
+@item @var{pvar} .ASSIGNC "@var{str}"
+Define a variable with a string value, for use during preprocessing.
+You can redefine variables with @code{.ASSIGNC} at any time.
+
+@item @var{pvar} .REG (@var{register})
+Use @code{.REG} to define a variable that represents a register.  In
+particular, @var{register} is @emph{not evaluated} as an expression.
+You may use @code{.REG} at will to redefine register variables.
+@end ftable
+
+All these directives accept the variable name in the ``label'' position,
+that is at the left margin.  You may specify a colon after the variable
+name if you wish; the first example above could have started @samp{eg:}
+with the same effect.
+
+@c pagebreak makes for better aesthetics---ensures macro and expansion together
+@page
+@node Macros
+@section Defining your own directives
+
+The commands @code{.MACRO} and @code{.ENDM} allow you to define macros
+that generate assembly output.  You can use these macros with a syntax
+similar to built-in @sc{gasp} or assembler directives.  For example,
+this definition specifies a macro @code{SUM} that adds together a range of
+consecutive registers:
+
+@cartouche
+@example
+        .MACRO  SUM FROM=0, TO=9
+        ! \FROM \TO
+        mov     r\FROM,r10
+COUNT   .ASSIGNA        \FROM+1
+        .AWHILE \&COUNT LE \TO
+        add     r\&COUNT,r10
+COUNT   .ASSIGNA        \&COUNT+1
+        .AENDW
+        .ENDM
+@end example
+@end cartouche
+
+@noindent
+With that definition, @samp{SUM 0,5} generates this assembly output:
+
+@cartouche
+@example
+        ! 0 5
+        mov     r0,r10
+        add     r1,r10
+        add     r2,r10
+        add     r3,r10
+        add     r4,r10
+        add     r5,r10
+@end example
+@end cartouche
+
+@ftable @code
+@item .MACRO @var{macname}
+@itemx .MACRO @var{macname} @var{macargs} @dots{}
+Begin the definition of a macro called @var{macname}.  If your macro
+definition requires arguments, specify their names after the macro name,
+separated by commas or spaces.  You can supply a default value for any
+macro argument by following the name with @samp{=@var{deflt}}.  For
+example, these are all valid @code{.MACRO} statements:
+
+@table @code
+@item .MACRO COMM
+Begin the definition of a macro called @code{COMM}, which takes no
+arguments.
+
+@item .MACRO PLUS1 P, P1
+@itemx .MACRO PLUS1 P P1
+Either statement begins the definition of a macro called @code{PLUS1},
+which takes two arguments; within the macro definition, write
+@samp{\P} or @samp{\P1} to evaluate the arguments.
+
+@item .MACRO RESERVE_STR P1=0 P2
+Begin the definition of a macro called @code{RESERVE_STR}, with two
+arguments.  The first argument has a default value, but not the second.
+After the definition is complete, you can call the macro either as
+@samp{RESERVE_STR @var{a},@var{b}} (with @samp{\P1} evaluating to
+@var{a} and @samp{\P2} evaluating to @var{b}), or as @samp{RESERVE_STR
+,@var{b}} (with @samp{\P1} evaluating as the default, in this case
+@samp{0}, and @samp{\P2} evaluating to @var{b}).
+@end table
+
+When you call a macro, you can specify the argument values either by
+position, or by keyword.  For example, @samp{SUM 9,17} is equivalent to
+@samp{SUM TO=17, FROM=9}.  Macro arguments are preprocessor variables
+similar to the variables you define with @samp{.ASSIGNA} or
+@samp{.ASSIGNC}; in particular, you can use them in conditionals or for
+loop control.  (The only difference is the prefix you write to evaluate
+the variable: for a macro argument, write @samp{\@var{argname}}, but for
+a preprocessor variable, write @samp{\&@var{varname}}.)
+
+@item @var{name} .MACRO
+@itemx @var{name} .MACRO ( @var{macargs} @dots{} )
+@c FIXME check: I think no error _and_ no args recognized if I use form
+@c       NAME  .MACRO   ARG ARG
+An alternative form of introducing a macro definition: specify the macro
+name in the label position, and the arguments (if any) between
+parentheses after the name.  Defaulting rules and usage work the same
+way as for the other macro definition syntax.
+
+@item .ENDM
+Mark the end of a macro definition.
+
+@item .EXITM
+Exit early from the current macro definition, @code{.AREPEAT} loop, or
+@code{.AWHILE} loop.
+
+@cindex number of macros executed
+@cindex macros, count executed
+@item \@@
+@sc{gasp} maintains a counter of how many macros it has
+executed in this pseudo-variable; you can copy that number to your
+output with @samp{\@@}, but @emph{only within a macro definition}.
+
+@item LOCAL @var{name} [ , @dots{} ]
+@emph{Warning: @code{LOCAL} is only available if you select ``alternate
+macro syntax'' with @samp{-a} or @samp{--alternate}.}  @xref{Alternate,,
+Alternate macro syntax}.
+
+Generate a string replacement for each of the @var{name} arguments, and
+replace any instances of @var{name} in each macro expansion.  The
+replacement string is unique in the assembly, and different for each
+separate macro expansion.  @code{LOCAL} allows you to write macros that
+define symbols, without fear of conflict between separate macro expansions.
+@end ftable
+
+@node Data
+@section Data output
+
+In assembly code, you often need to specify working areas of memory;
+depending on the application, you may want to initialize such memory or
+not.  @sc{gasp} provides preprocessor directives to help you avoid
+repetitive coding for both purposes.
+
+You can use labels as usual to mark the data areas.
+
+@menu
+* Initialized::
+* Uninitialized::
+@end menu
+
+@node Initialized
+@subsection Initialized data
+
+These are the @sc{gasp} directives for initialized data, and the standard
+@sc{gnu} assembler directives they expand to:
+
+@ftable @code
+@item .DATA @var{expr}, @var{expr}, @dots{}
+@itemx .DATA.B @var{expr}, @var{expr}, @dots{}
+@itemx .DATA.W @var{expr}, @var{expr}, @dots{}
+@itemx .DATA.L @var{expr}, @var{expr}, @dots{}
+Evaluate arithmetic expressions @var{expr}, and emit the corresponding
+@code{as} directive (labelled with @var{lab}).  The unqualified
+@code{.DATA} emits @samp{.long}; @code{.DATA.B} emits @samp{.byte};
+@code{.DATA.W} emits @samp{.short}; and @code{.DATA.L} emits
+@samp{.long}.
+
+For example, @samp{foo .DATA 1,2,3} emits @samp{foo: .long 1,2,3}.
+
+@item .DATAB @var{repeat}, @var{expr}
+@itemx .DATAB.B @var{repeat}, @var{expr}
+@itemx .DATAB.W @var{repeat}, @var{expr}
+@itemx .DATAB.L @var{repeat}, @var{expr}
+@c FIXME! Looks like gasp accepts and ignores args after 2nd.
+Make @code{as} emit @var{repeat} copies of the value of the expression
+@var{expr} (using the @code{as} directive @code{.fill}).
+@samp{.DATAB.B} repeats one-byte values; @samp{.DATAB.W} repeats
+two-byte values; and @samp{.DATAB.L} repeats four-byte values.
+@samp{.DATAB} without a suffix repeats four-byte values, just like
+@samp{.DATAB.L}.
+
+@c FIXME! Allowing zero might be useful for edge conditions in macros.
+@var{repeat} must be an absolute expression with a positive value.
+
+@item .SDATA "@var{str}" @dots{}
+String data.  Emits a concatenation of bytes, precisely as you specify
+them (in particular, @emph{nothing is added to mark the end} of the
+string).  @xref{Constants,, String and numeric constants}, for details
+about how to write strings.  @code{.SDATA} concatenates multiple
+arguments, making it easy to switch between string representations.  You
+can use commas to separate the individual arguments for clarity, if you
+choose.
+
+@item .SDATAB @var{repeat}, "@var{str}" @dots{}
+Repeated string data.  The first argument specifies how many copies of
+the string to emit; the remaining arguments specify the string, in the
+same way as the arguments to @code{.SDATA}.
+
+@item .SDATAZ "@var{str}" @dots{}
+Zero-terminated string data.  Just like @code{.SDATA}, except that
+@code{.SDATAZ} writes a zero byte at the end of the string.
+
+@item .SDATAC "@var{str}" @dots{}
+Count-prefixed string data.  Just like @code{.SDATA}, except that
+@sc{gasp} precedes the string with a leading one-byte count.  For
+example, @samp{.SDATAC "HI"} generates @samp{.byte 2,72,73}.  Since the
+count field is only one byte, you can only use @code{.SDATAC} for
+strings less than 256 bytes in length.
+@end ftable
+
+@node Uninitialized
+@subsection Uninitialized data
+
+@c FIXME!  .space different on some platforms, notably HPPA.  Config?
+Use the @code{.RES}, @code{.SRES}, @code{.SRESC}, and @code{.SRESZ}
+directives to reserve memory and leave it uninitialized.  @sc{gasp}
+resolves these directives to appropriate calls of the @sc{gnu}
+@code{as} @code{.space} directive.
+
+@ftable @code
+@item .RES @var{count}
+@itemx .RES.B @var{count}
+@itemx .RES.W @var{count}
+@itemx .RES.L @var{count}
+Reserve room for @var{count} uninitialized elements of data.  The
+suffix specifies the size of each element: @code{.RES.B} reserves
+@var{count} bytes, @code{.RES.W} reserves @var{count} pairs of bytes,
+and @code{.RES.L} reserves @var{count} quartets.  @code{.RES} without a
+suffix is equivalent to @code{.RES.L}.
+
+@item .SRES @var{count}
+@itemx .SRES.B @var{count}
+@itemx .SRES.W @var{count}
+@itemx .SRES.L @var{count}
+@c FIXME!  This is boring.  Shouldn't it at least have a different
+@c         default size?  (e.g. the "S" suggests "string", for which .B
+@c         would be more appropriate)
+@code{.SRES} is a synonym for @samp{.RES}.
+
+@item .SRESC @var{count}
+@itemx .SRESC.B @var{count}
+@itemx .SRESC.W @var{count}
+@itemx .SRESC.L @var{count}
+Like @code{.SRES}, but reserves space for @code{@var{count}+1} elements.
+
+@item .SRESZ @var{count}
+@itemx .SRESZ.B @var{count}
+@itemx .SRESZ.W @var{count}
+@itemx .SRESZ.L @var{count}
+Like @code{.SRES}, but reserves space for @code{@var{count}+1} elements.
+@end ftable
+
+@node Listings
+@section Assembly listing control
+
+The @sc{gasp} listing-control directives correspond to
+related @sc{gnu} @code{as} directives.
+
+@ftable @code
+@item .PRINT LIST
+@itemx .PRINT NOLIST
+Print control.  This directive emits the @sc{gnu} @code{as} directive
+@code{.list} or @code{.nolist}, according to its argument.  @xref{List,,
+@code{.list}, as.info, Using as}, for details on how these directives
+interact.
+
+@item .FORM LIN=@var{ln}
+@itemx .FORM COL=@var{cols}
+@itemx .FORM LIN=@var{ln} COL=@var{cols}
+Specify the page size for assembly listings: @var{ln} represents the
+number of lines, and @var{cols} the number of columns.  You may specify
+either page dimension independently, or both together.  If you do not
+specify the number of lines, @sc{gasp} assumes 60 lines; if you do not
+specify the number of columns, @sc{gasp} assumes 132 columns.
+(Any values you may have specified in previous instances of @code{.FORM}
+do @emph{not} carry over as defaults.)  Emits the @code{.psize}
+assembler directive.
+
+@item .HEADING @var{string}
+Specify @var{string} as the title of your assembly listings.  Emits
+@samp{.title "@var{string}"}.
+
+@item .PAGE
+Force a new page in assembly listings.  Emits @samp{.eject}.
+@end ftable
+
+@node Other Commands
+@section Miscellaneous commands
+
+@ftable @code
+@item .ALTERNATE
+Use the alternate macro syntax henceforth in the assembly.
+@xref{Alternate,, Alternate macro syntax}.
+
+@item .ORG
+@c FIXME!  This is very strange, since _GAS_ understands .org
+This command is recognized, but not yet implemented.  @sc{gasp}
+generates an error message for programs that use @code{.ORG}.
+
+@item .RADIX @var{s}
+@c FIXME no test cases in testsuite/gasp
+@sc{gasp} understands numbers in any of base two, eight, ten, or
+sixteen.  You can encode the base explicitly in any numeric constant
+(@pxref{Constants,, String and numeric constants}).  If you write
+numbers without an explicit indication of the base, the most recent
+@samp{.RADIX @var{s}} command determines how they are interpreted.
+@var{s} is a single letter, one of the following:
+
+@table @code
+@item .RADIX B
+Base 2.
+
+@item .RADIX Q
+Base 8.
+
+@item .RADIX D
+Base 10.  This is the original default radix.
+
+@item .RADIX H
+Base 16.
+@end table
+
+You may specify the argument @var{s} in lower case (any of @samp{bqdh})
+with the same effects.
+
+@item .EXPORT @var{name}
+@itemx .GLOBAL @var{name}
+@c FIXME! No test cases in testsuite/gasp
+Declare @var{name} global (emits @samp{.global @var{name}}).  The two
+directives are synonymous.
+
+@item .PROGRAM
+No effect: @sc{gasp} accepts this directive, and silently ignores it.
+
+@item .END
+Mark end of each preprocessor file.  @sc{gasp} issues a warning if it
+reaches end of file without seeing this command.
+
+@item .INCLUDE "@var{str}"
+Preprocess the file named by @var{str}, as if its contents appeared
+where the @code{.INCLUDE} directive does.  @sc{gasp} imposes a maximum
+limit of 30 stacked include files, as a sanity check.
+@c FIXME!  Why is include depth not affected by -u?
+
+@item .ALIGN @var{size}
+@c FIXME! Why is this not utterly pointless?
+Evaluate the absolute expression @var{size}, and emit the assembly
+instruction @samp{.align @var{size}} using the result.
+@end ftable
+
+@node Syntax Details
+@section Details of the GASP syntax
+
+Since @sc{gasp} is meant to work with assembly code, its statement
+syntax has no surprises for the assembly programmer.
+
+@cindex whitespace
+@emph{Whitespace} (blanks or tabs; @emph{not} newline) is partially
+significant, in that it delimits up to three fields in a line.  The
+amount of whitespace does not matter; you may line up fields in separate
+lines if you wish, but @sc{gasp} does not require that.
+
+@cindex fields of @sc{gasp} source line
+@cindex label field
+The @emph{first field}, an optional @dfn{label}, must be flush left in a
+line (with no leading whitespace) if it appears at all.  You may use a
+colon after the label if you wish; @sc{gasp} neither requires the colon
+nor objects to it (but will not include it as part of the label name).
+
+@cindex directive field
+The @emph{second field}, which must appear after some whitespace,
+contains a @sc{gasp} or assembly @dfn{directive}.
+
+@cindex argument fields
+Any @emph{further fields} on a line are @dfn{arguments} to the
+directive; you can separate them from one another using either commas or
+whitespace.
+
+@menu
+* Markers::
+* Constants::
+* Symbols::
+* Expressions::
+* String Builtins::
+@end menu
+
+@node Markers
+@subsection Special syntactic markers
+
+@sc{gasp} recognizes a few special markers: to delimit comments, to
+continue a statement on the next line, to separate symbols from other
+characters, and to copy text to the output literally.  (One other
+special marker, @samp{\@@}, works only within macro definitions;
+@pxref{Macros,, Defining your own directives}.)
+
+@cindex comments
+The trailing part of any @sc{gasp} source line may be a @dfn{comment}.
+A comment begins with the first unquoted comment character (@samp{!} by
+default), or an escaped or doubled comment character (@samp{\!} or
+@samp{!!} by default), and extends to the end of a line.  You can
+specify what comment character to use with the @samp{-c} option
+(@pxref{Invoking GASP,, Command Line Options}).  The two kinds of
+comment markers lead to slightly different treatment:
+
+@table @code
+@item !
+A single, un-escaped comment character generates an assembly comment in
+the @sc{gasp} output.  @sc{gasp} evaluates any preprocessor variables
+(macro arguments, or variables defined with @code{.ASSIGNA} or
+@code{.ASSIGNC}) present.  For example, a macro that begins like this
+
+@example
+        .MACRO  SUM FROM=0, TO=9
+        ! \FROM \TO
+@end example
+
+@noindent
+issues as the first line of output a comment that records the
+values you used to call the macro.
+
+@c comments, preprocessor-only
+@c preprocessor-only comments
+@c GASP-only comments
+@item \!
+@itemx !!
+Either an escaped comment character, or a double comment character,
+marks a @sc{gasp} source comment.  @sc{gasp} does not copy such comments
+to the assembly output.
+@end table
+
+@cindex continuation character
+@kindex +
+To @emph{continue a statement} on the next line of the file, begin the
+second line with the character @samp{+}.
+
+@cindex literal copy to output
+@cindex copying literally to output
+@cindex preprocessing, avoiding
+@cindex avoiding preprocessing
+Occasionally you may want to prevent @sc{gasp} from preprocessing some
+particular bit of text.  To @emph{copy literally} from the @sc{gasp}
+source to its output, place @samp{\(} before the string to copy, and
+@samp{)} at the end.  For example, write @samp{\(\!)} if you need the
+characters @samp{\!} in your assembly output.
+
+@cindex symbol separator
+@cindex text, separating from symbols
+@cindex symbols, separating from text
+To @emph{separate a preprocessor variable} from text to appear
+immediately after its value, write a single quote (@code{'}).  For
+example, @samp{.SDATA "\P'1"} writes a string built by concatenating the
+value of @code{P} and the digit @samp{1}.  (You cannot achieve this by
+writing just @samp{\P1}, since @samp{P1} is itself a valid name for a
+preprocessor variable.)
+
+@node Constants
+@subsection String and numeric constants
+
+There are two ways of writing @dfn{string constants} in @sc{gasp}: as
+literal text, and by numeric byte value.  Specify a string literal
+between double quotes (@code{"@var{str}"}).  Specify an individual
+numeric byte value as an absolute expression between angle brackets
+(@code{<@var{expr}>}.  Directives that output strings allow you to
+specify any number of either kind of value, in whatever order is
+convenient, and concatenate the result.  (Alternate syntax mode
+introduces a number of alternative string notations; @pxref{Alternate,,
+Alternate macro syntax}.)
+
+@c Details of numeric notation, e.g. base prefixes
+You can write @dfn{numeric constants} either in a specific base, or in
+whatever base is currently selected (either 10, or selected by the most
+recent @code{.RADIX}).
+
+To write a number in a @emph{specific base}, use the pattern
+@code{@var{s}'@var{ddd}}: a base specifier character @var{s}, followed
+by a single quote followed by digits @var{ddd}.  The base specifier
+character matches those you can specify with @code{.RADIX}: @samp{B} for
+base 2, @samp{Q} for base 8, @samp{D} for base 10, and @samp{H} for base
+16.  (You can write this character in lower case if you prefer.)
+
+You can write floating point constants using the same syntax recognised
+by GAS @ref{Flonums,,Flonums,as,The GNU Assembler.}.  A constraint is
+that these constants will be interpreted as decimal values irrespective
+of the currently selected base.
+
+@c FIXME! What are rules for recognizing number in deflt base?  Whatever
+@c        is left over after parsing other things??
+
+@node Symbols
+@subsection Symbols
+
+@sc{gasp} recognizes symbol names that start with any alphabetic character,
+@samp{_}, or @samp{$}, and continue with any of the same characters or
+with digits.  Label names follow the same rules.
+
+@node Expressions
+@subsection Arithmetic expressions in GASP
+
+@cindex absolute expressions
+@cindex relocatable expressions
+There are two kinds of expressions, depending on their result:
+@dfn{absolute} expressions, which resolve to a constant (that is, they
+do not involve any values unknown to @sc{gasp}), and @dfn{relocatable}
+expressions, which must reduce to the form
+
+@example
+@var{addsym}+@var{const}-@var{subsym}
+@end example
+
+@noindent
+where @var{addsym} and @var{subsym} are assembly symbols of unknown
+value, and @var{const} is a constant.
+
+Arithmetic for @sc{gasp} expressions follows very similar rules to C.
+You can use parentheses to change precedence; otherwise, arithmetic
+primitives have decreasing precedence in the order of the following
+list.
+
+@enumerate
+@item
+Single-argument @code{+} (identity), @code{-} (arithmetic opposite), or
+@code{~} (bitwise negation).  @emph{The argument must be an absolute
+expression.}
+
+@item
+@code{*} (multiplication) and @code{/} (division).  @emph{Both arguments
+must be absolute expressions.}
+
+@item
+@code{+} (addition) and @code{-} (subtraction).  @emph{At least one argument
+must be absolute.}
+@c FIXME!  Actually, subtraction doesn't check for this.
+
+@item
+@code{&} (bitwise and).  @emph{Both arguments must be absolute.}
+
+@item
+@c FIXME!  I agree ~ is a better notation than ^ for xor, but is the
+@c         improvement worth differing from C?
+@code{|} (bitwise or) and @code{~} (bitwise exclusive or; @code{^} in
+C).  @emph{Both arguments must be absolute.}
+@end enumerate
+
+@node String Builtins
+@subsection String primitives
+
+You can use these primitives to manipulate strings (in the argument
+field of @sc{gasp} statements):
+
+@ftable @code
+@item .LEN("@var{str}")
+Calculate the length of string @code{"@var{str}"}, as an absolute
+expression.  For example, @samp{.RES.B .LEN("sample")} reserves six
+bytes of memory.
+
+@item .INSTR("@var{string}", "@var{seg}", @var{ix})
+Search for the first occurrence of @var{seg} after position @var{ix} of
+@var{string}.  For example, @samp{.INSTR("ABCDEFG", "CDE", 0)} evaluates
+to the absolute result @code{2}.
+
+The result is @code{-1} if @var{seg} does not occur in @var{string}
+after position @var{ix}.
+
+@item .SUBSTR("@var{string}",@var{start},@var{len})
+The substring of @var{string} beginning at byte number @var{start} and
+extending for @var{len} bytes.
+@end ftable
+
+@node Alternate
+@section Alternate macro syntax
+
+If you specify @samp{-a} or @samp{--alternate} on the @sc{gasp} command
+line, the preprocessor uses somewhat different syntax.  This syntax is
+reminiscent of the syntax of Phar Lap macro assembler, but it
+is @emph{not} meant to be a full emulation of Phar Lap or similar
+assemblers.  In particular, @sc{gasp} does not support directives such
+as @code{DB} and @code{IRP}, even in alternate syntax mode.
+
+In particular, @samp{-a} (or @samp{--alternate}) elicits these
+differences:
+
+@table @emph
+@item Preprocessor directives
+You can use @sc{gasp} preprocessor directives without a leading @samp{.}
+dot.  For example, you can write @samp{SDATA} with the same effect as
+@samp{.SDATA}.
+
+@item LOCAL
+One additional directive, @code{LOCAL}, is available.  @xref{Macros,,
+Defining your own directives}, for an explanation of how to use
+@code{LOCAL}.
+
+@need 2000
+@item String delimiters
+You can write strings delimited in these other ways besides
+@code{"@var{string}"}:
+
+@table @code
+@item '@var{string}'
+You can delimit strings with single-quote charaters.
+
+@item <@var{string}>
+You can delimit strings with matching angle brackets.
+@end table
+
+@item single-character string escape
+To include any single character literally in a string (even if the
+character would otherwise have some special meaning), you can prefix the
+character with @samp{!} (an exclamation mark).  For example, you can
+write @samp{<4.3 !> 5.4!!>} to get the literal text @samp{4.3 > 5.4!}.
+
+@item Expression results as strings
+You can write @samp{%@var{expr}} to evaluate the expression @var{expr}
+and use the result as a string.  
+@end table
+
+@node GNU Free Documentation License
+@chapter GNU Free Documentation License
+
+                GNU Free Documentation License
+                
+                   Version 1.1, March 2000
+
+ Copyright (C) 2000  Free Software Foundation, Inc.
+  59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+     
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+
+0. PREAMBLE
+
+The purpose of this License is to make a manual, textbook, or other
+written document "free" in the sense of freedom: to assure everyone
+the effective freedom to copy and redistribute it, with or without
+modifying it, either commercially or noncommercially.  Secondarily,
+this License preserves for the author and publisher a way to get
+credit for their work, while not being considered responsible for
+modifications made by others.
+
+This License is a kind of "copyleft", which means that derivative
+works of the document must themselves be free in the same sense.  It
+complements the GNU General Public License, which is a copyleft
+license designed for free software.
+
+We have designed this License in order to use it for manuals for free
+software, because free software needs free documentation: a free
+program should come with manuals providing the same freedoms that the
+software does.  But this License is not limited to software manuals;
+it can be used for any textual work, regardless of subject matter or
+whether it is published as a printed book.  We recommend this License
+principally for works whose purpose is instruction or reference.
+
+
+1. APPLICABILITY AND DEFINITIONS
+
+This License applies to any manual or other work that contains a
+notice placed by the copyright holder saying it can be distributed
+under the terms of this License.  The "Document", below, refers to any
+such manual or work.  Any member of the public is a licensee, and is
+addressed as "you".
+
+A "Modified Version" of the Document means any work containing the
+Document or a portion of it, either copied verbatim, or with
+modifications and/or translated into another language.
+
+A "Secondary Section" is a named appendix or a front-matter section of
+the Document that deals exclusively with the relationship of the
+publishers or authors of the Document to the Document's overall subject
+(or to related matters) and contains nothing that could fall directly
+within that overall subject.  (For example, if the Document is in part a
+textbook of mathematics, a Secondary Section may not explain any
+mathematics.)  The relationship could be a matter of historical
+connection with the subject or with related matters, or of legal,
+commercial, philosophical, ethical or political position regarding
+them.
+
+The "Invariant Sections" are certain Secondary Sections whose titles
+are designated, as being those of Invariant Sections, in the notice
+that says that the Document is released under this License.
+
+The "Cover Texts" are certain short passages of text that are listed,
+as Front-Cover Texts or Back-Cover Texts, in the notice that says that
+the Document is released under this License.
+
+A "Transparent" copy of the Document means a machine-readable copy,
+represented in a format whose specification is available to the
+general public, whose contents can be viewed and edited directly and
+straightforwardly with generic text editors or (for images composed of
+pixels) generic paint programs or (for drawings) some widely available
+drawing editor, and that is suitable for input to text formatters or
+for automatic translation to a variety of formats suitable for input
+to text formatters.  A copy made in an otherwise Transparent file
+format whose markup has been designed to thwart or discourage
+subsequent modification by readers is not Transparent.  A copy that is
+not "Transparent" is called "Opaque".
+
+Examples of suitable formats for Transparent copies include plain
+ASCII without markup, Texinfo input format, LaTeX input format, SGML
+or XML using a publicly available DTD, and standard-conforming simple
+HTML designed for human modification.  Opaque formats include
+PostScript, PDF, proprietary formats that can be read and edited only
+by proprietary word processors, SGML or XML for which the DTD and/or
+processing tools are not generally available, and the
+machine-generated HTML produced by some word processors for output
+purposes only.
+
+The "Title Page" means, for a printed book, the title page itself,
+plus such following pages as are needed to hold, legibly, the material
+this License requires to appear in the title page.  For works in
+formats which do not have any title page as such, "Title Page" means
+the text near the most prominent appearance of the work's title,
+preceding the beginning of the body of the text.
+
+
+2. VERBATIM COPYING
+
+You may copy and distribute the Document in any medium, either
+commercially or noncommercially, provided that this License, the
+copyright notices, and the license notice saying this License applies
+to the Document are reproduced in all copies, and that you add no other
+conditions whatsoever to those of this License.  You may not use
+technical measures to obstruct or control the reading or further
+copying of the copies you make or distribute.  However, you may accept
+compensation in exchange for copies.  If you distribute a large enough
+number of copies you must also follow the conditions in section 3.
+
+You may also lend copies, under the same conditions stated above, and
+you may publicly display copies.
+
+
+3. COPYING IN QUANTITY
+
+If you publish printed copies of the Document numbering more than 100,
+and the Document's license notice requires Cover Texts, you must enclose
+the copies in covers that carry, clearly and legibly, all these Cover
+Texts: Front-Cover Texts on the front cover, and Back-Cover Texts on
+the back cover.  Both covers must also clearly and legibly identify
+you as the publisher of these copies.  The front cover must present
+the full title with all words of the title equally prominent and
+visible.  You may add other material on the covers in addition.
+Copying with changes limited to the covers, as long as they preserve
+the title of the Document and satisfy these conditions, can be treated
+as verbatim copying in other respects.
+
+If the required texts for either cover are too voluminous to fit
+legibly, you should put the first ones listed (as many as fit
+reasonably) on the actual cover, and continue the rest onto adjacent
+pages.
+
+If you publish or distribute Opaque copies of the Document numbering
+more than 100, you must either include a machine-readable Transparent
+copy along with each Opaque copy, or state in or with each Opaque copy
+a publicly-accessible computer-network location containing a complete
+Transparent copy of the Document, free of added material, which the
+general network-using public has access to download anonymously at no
+charge using public-standard network protocols.  If you use the latter
+option, you must take reasonably prudent steps, when you begin
+distribution of Opaque copies in quantity, to ensure that this
+Transparent copy will remain thus accessible at the stated location
+until at least one year after the last time you distribute an Opaque
+copy (directly or through your agents or retailers) of that edition to
+the public.
+
+It is requested, but not required, that you contact the authors of the
+Document well before redistributing any large number of copies, to give
+them a chance to provide you with an updated version of the Document.
+
+
+4. MODIFICATIONS
+
+You may copy and distribute a Modified Version of the Document under
+the conditions of sections 2 and 3 above, provided that you release
+the Modified Version under precisely this License, with the Modified
+Version filling the role of the Document, thus licensing distribution
+and modification of the Modified Version to whoever possesses a copy
+of it.  In addition, you must do these things in the Modified Version:
+
+A. Use in the Title Page (and on the covers, if any) a title distinct
+   from that of the Document, and from those of previous versions
+   (which should, if there were any, be listed in the History section
+   of the Document).  You may use the same title as a previous version
+   if the original publisher of that version gives permission.
+B. List on the Title Page, as authors, one or more persons or entities
+   responsible for authorship of the modifications in the Modified
+   Version, together with at least five of the principal authors of the
+   Document (all of its principal authors, if it has less than five).
+C. State on the Title page the name of the publisher of the
+   Modified Version, as the publisher.
+D. Preserve all the copyright notices of the Document.
+E. Add an appropriate copyright notice for your modifications
+   adjacent to the other copyright notices.
+F. Include, immediately after the copyright notices, a license notice
+   giving the public permission to use the Modified Version under the
+   terms of this License, in the form shown in the Addendum below.
+G. Preserve in that license notice the full lists of Invariant Sections
+   and required Cover Texts given in the Document's license notice.
+H. Include an unaltered copy of this License.
+I. Preserve the section entitled "History", and its title, and add to
+   it an item stating at least the title, year, new authors, and
+   publisher of the Modified Version as given on the Title Page.  If
+   there is no section entitled "History" in the Document, create one
+   stating the title, year, authors, and publisher of the Document as
+   given on its Title Page, then add an item describing the Modified
+   Version as stated in the previous sentence.
+J. Preserve the network location, if any, given in the Document for
+   public access to a Transparent copy of the Document, and likewise
+   the network locations given in the Document for previous versions
+   it was based on.  These may be placed in the "History" section.
+   You may omit a network location for a work that was published at
+   least four years before the Document itself, or if the original
+   publisher of the version it refers to gives permission.
+K. In any section entitled "Acknowledgements" or "Dedications",
+   preserve the section's title, and preserve in the section all the
+   substance and tone of each of the contributor acknowledgements
+   and/or dedications given therein.
+L. Preserve all the Invariant Sections of the Document,
+   unaltered in their text and in their titles.  Section numbers
+   or the equivalent are not considered part of the section titles.
+M. Delete any section entitled "Endorsements".  Such a section
+   may not be included in the Modified Version.
+N. Do not retitle any existing section as "Endorsements"
+   or to conflict in title with any Invariant Section.
+
+If the Modified Version includes new front-matter sections or
+appendices that qualify as Secondary Sections and contain no material
+copied from the Document, you may at your option designate some or all
+of these sections as invariant.  To do this, add their titles to the
+list of Invariant Sections in the Modified Version's license notice.
+These titles must be distinct from any other section titles.
+
+You may add a section entitled "Endorsements", provided it contains
+nothing but endorsements of your Modified Version by various
+parties--for example, statements of peer review or that the text has
+been approved by an organization as the authoritative definition of a
+standard.
+
+You may add a passage of up to five words as a Front-Cover Text, and a
+passage of up to 25 words as a Back-Cover Text, to the end of the list
+of Cover Texts in the Modified Version.  Only one passage of
+Front-Cover Text and one of Back-Cover Text may be added by (or
+through arrangements made by) any one entity.  If the Document already
+includes a cover text for the same cover, previously added by you or
+by arrangement made by the same entity you are acting on behalf of,
+you may not add another; but you may replace the old one, on explicit
+permission from the previous publisher that added the old one.
+
+The author(s) and publisher(s) of the Document do not by this License
+give permission to use their names for publicity for or to assert or
+imply endorsement of any Modified Version.
+
+
+5. COMBINING DOCUMENTS
+
+You may combine the Document with other documents released under this
+License, under the terms defined in section 4 above for modified
+versions, provided that you include in the combination all of the
+Invariant Sections of all of the original documents, unmodified, and
+list them all as Invariant Sections of your combined work in its
+license notice.
+
+The combined work need only contain one copy of this License, and
+multiple identical Invariant Sections may be replaced with a single
+copy.  If there are multiple Invariant Sections with the same name but
+different contents, make the title of each such section unique by
+adding at the end of it, in parentheses, the name of the original
+author or publisher of that section if known, or else a unique number.
+Make the same adjustment to the section titles in the list of
+Invariant Sections in the license notice of the combined work.
+
+In the combination, you must combine any sections entitled "History"
+in the various original documents, forming one section entitled
+"History"; likewise combine any sections entitled "Acknowledgements",
+and any sections entitled "Dedications".  You must delete all sections
+entitled "Endorsements."
+
+
+6. COLLECTIONS OF DOCUMENTS
+
+You may make a collection consisting of the Document and other documents
+released under this License, and replace the individual copies of this
+License in the various documents with a single copy that is included in
+the collection, provided that you follow the rules of this License for
+verbatim copying of each of the documents in all other respects.
+
+You may extract a single document from such a collection, and distribute
+it individually under this License, provided you insert a copy of this
+License into the extracted document, and follow this License in all
+other respects regarding verbatim copying of that document.
+
+
+7. AGGREGATION WITH INDEPENDENT WORKS
+
+A compilation of the Document or its derivatives with other separate
+and independent documents or works, in or on a volume of a storage or
+distribution medium, does not as a whole count as a Modified Version
+of the Document, provided no compilation copyright is claimed for the
+compilation.  Such a compilation is called an "aggregate", and this
+License does not apply to the other self-contained works thus compiled
+with the Document, on account of their being thus compiled, if they
+are not themselves derivative works of the Document.
+
+If the Cover Text requirement of section 3 is applicable to these
+copies of the Document, then if the Document is less than one quarter
+of the entire aggregate, the Document's Cover Texts may be placed on
+covers that surround only the Document within the aggregate.
+Otherwise they must appear on covers around the whole aggregate.
+
+
+8. TRANSLATION
+
+Translation is considered a kind of modification, so you may
+distribute translations of the Document under the terms of section 4.
+Replacing Invariant Sections with translations requires special
+permission from their copyright holders, but you may include
+translations of some or all Invariant Sections in addition to the
+original versions of these Invariant Sections.  You may include a
+translation of this License provided that you also include the
+original English version of this License.  In case of a disagreement
+between the translation and the original English version of this
+License, the original English version will prevail.
+
+
+9. TERMINATION
+
+You may not copy, modify, sublicense, or distribute the Document except
+as expressly provided for under this License.  Any other attempt to
+copy, modify, sublicense or distribute the Document is void, and will
+automatically terminate your rights under this License.  However,
+parties who have received copies, or rights, from you under this
+License will not have their licenses terminated so long as such
+parties remain in full compliance.
+
+
+10. FUTURE REVISIONS OF THIS LICENSE
+
+The Free Software Foundation may publish new, revised versions
+of the GNU Free Documentation License from time to time.  Such new
+versions will be similar in spirit to the present version, but may
+differ in detail to address new problems or concerns.  See
+http://www.gnu.org/copyleft/.
+
+Each version of the License is given a distinguishing version number.
+If the Document specifies that a particular numbered version of this
+License "or any later version" applies to it, you have the option of
+following the terms and conditions either of that specified version or
+of any later version that has been published (not as a draft) by the
+Free Software Foundation.  If the Document does not specify a version
+number of this License, you may choose any version ever published (not
+as a draft) by the Free Software Foundation.
+
+
+ADDENDUM: How to use this License for your documents
+
+To use this License in a document you have written, include a copy of
+the License in the document and put the following copyright and
+license notices just after the title page:
+
+@smallexample
+    Copyright (c)  YEAR  YOUR NAME.
+    Permission is granted to copy, distribute and/or modify this document
+    under the terms of the GNU Free Documentation License, Version 1.1
+    or any later version published by the Free Software Foundation;
+    with the Invariant Sections being LIST THEIR TITLES, with the
+    Front-Cover Texts being LIST, and with the Back-Cover Texts being LIST.
+    A copy of the license is included in the section entitled "GNU
+    Free Documentation License".
+@end smallexample
+
+If you have no Invariant Sections, write "with no Invariant Sections"
+instead of saying which ones are invariant.  If you have no
+Front-Cover Texts, write "no Front-Cover Texts" instead of
+"Front-Cover Texts being LIST"; likewise for Back-Cover Texts.
+
+If your document contains nontrivial examples of program code, we
+recommend releasing these examples in parallel under your choice of
+free software license, such as the GNU General Public License,
+to permit their use in free software.
+
+@node Index
+@unnumbered Index
+
+@printindex cp
+
+@contents
+@bye
diff -urNbB binutils-2.16.91.0.2.org/gas/gasp.c binutils-2.16.91.0.2/gas/gasp.c
--- binutils-2.16.91.0.2.org/gas/gasp.c	1970-01-01 00:00:00.000000000 +0000
+++ binutils-2.16.91.0.2/gas/gasp.c	2005-07-21 18:31:04.000000000 +0000
@@ -0,0 +1,3983 @@
+/* gasp.c - Gnu assembler preprocessor main program.
+   Copyright 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002
+   Free Software Foundation, Inc.
+
+   Written by Steve and Judy Chamberlain of Cygnus Support,
+      sac@cygnus.com
+
+   This file is part of GASP, the GNU Assembler Preprocessor.
+
+   GASP is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation; either version 2, or (at your option)
+   any later version.
+
+   GASP is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with GASP; see the file COPYING.  If not, write to the Free
+   Software Foundation, 59 Temple Place - Suite 330, Boston, MA
+   02111-1307, USA.  */
+
+/*
+This program translates the input macros and stuff into a form
+suitable for gas to consume.
+
+  gasp [-sdhau] [-c char] [-o <outfile>] <infile>*
+
+  -s copy source to output
+  -c <char> comments are started with <char> instead of !
+  -u allow unreasonable stuff
+  -p print line numbers
+  -d print debugging stats
+  -s semi colons start comments
+  -a use alternate syntax
+     Pseudo ops can start with or without a .
+     Labels have to be in first column.
+  -I specify include dir
+    Macro arg parameters subsituted by name, don't need the &.
+     String can start with ' too.
+     Strings can be surrounded by <..>
+     A %<exp> in a string evaluates the expression
+     Literal char in a string with !
+*/
+
+#include "config.h"
+#include "bin-bugs.h"
+
+#include <assert.h>
+#include <stdio.h>
+#include <string.h>
+#include "getopt.h"
+
+#ifdef HAVE_STDLIB_H
+#include <stdlib.h>
+#endif
+
+#ifdef NEED_MALLOC_DECLARATION
+extern char *malloc ();
+#endif
+
+#include "ansidecl.h"
+#include "libiberty.h"
+#include "safe-ctype.h"
+#include "sb.h"
+#include "macro.h"
+#include "asintl.h"
+#include "xregex.h"
+
+char *program_version = "1.2";
+
+/* This is normally declared in as.h, but we don't include that.  We
+   need the function because other files linked with gasp.c might call
+   it.  */
+extern void as_abort PARAMS ((const char *, int, const char *));
+
+/* The default obstack chunk size.  If we set this to zero, the
+   obstack code will use whatever will fit in a 4096 byte block.  This
+   is used by the hash table code used by macro.c.  */
+int chunksize = 0;
+
+#define MAX_INCLUDES 30		/* Maximum include depth.  */
+#define MAX_REASONABLE 1000	/* Maximum number of expansions.  */
+
+int unreasonable;		/* -u on command line.  */
+int stats;			/* -d on command line.  */
+int print_line_number;		/* -p flag on command line.  */
+int copysource;			/* -c flag on command line.  */
+int warnings;			/* Number of WARNINGs generated so far.  */
+int errors;			/* Number of ERRORs generated so far.  */
+int fatals;			/* Number of fatal ERRORs generated so far (either 0 or 1).  */
+int alternate = 0;              /* -a on command line.  */
+int mri = 0;			/* -M on command line.  */
+char comment_char = '!';
+int radix = 10;			/* Default radix.  */
+
+int had_end; /* Seen .END.  */
+
+/* The output stream.  */
+FILE *outfile;
+
+/* The attributes of each character are stored as a bit pattern
+   chartype, which gives us quick tests.  */
+
+#define FIRSTBIT 1
+#define NEXTBIT  2
+#define SEPBIT   4
+#define WHITEBIT 8
+#define COMMENTBIT 16
+#define BASEBIT  32
+#define ISCOMMENTCHAR(x) (chartype[(unsigned char)(x)] & COMMENTBIT)
+#define ISFIRSTCHAR(x)  (chartype[(unsigned char)(x)] & FIRSTBIT)
+#define ISNEXTCHAR(x)   (chartype[(unsigned char)(x)] & NEXTBIT)
+#define ISSEP(x)        (chartype[(unsigned char)(x)] & SEPBIT)
+#define ISWHITE(x)      (chartype[(unsigned char)(x)] & WHITEBIT)
+#define ISBASE(x)       (chartype[(unsigned char)(x)] & BASEBIT)
+static char chartype[256];
+
+/* Conditional assembly uses the `ifstack'.  Each aif pushes another
+   entry onto the stack, and sets the on flag if it should.  The aelse
+   sets hadelse, and toggles on.  An aend pops a level.  We limit to
+   100 levels of nesting, not because we're facists pigs with read
+   only minds, but because more than 100 levels of nesting is probably
+   a bug in the user's macro structure.  */
+
+#define IFNESTING 100
+struct {
+  int on;			/* Is the level being output.  */
+  int hadelse;			/* Has an aelse been seen.  */
+} ifstack[IFNESTING];
+
+int ifi;
+
+/* The final and intermediate results of expression evaluation are kept in
+   exp_t's.  Note that a symbol is not an sb, but a pointer into the input
+   line.  It must be coped somewhere safe before the next line is read in.  */
+
+typedef struct {
+  char *name;
+  int len;
+} symbol;
+
+typedef struct {
+  int value;			/* Constant part.  */
+  symbol add_symbol;		/* Name part.  */
+  symbol sub_symbol;		/* Name part.  */
+} exp_t;
+
+/* Hashing is done in a pretty standard way.  A hash_table has a
+   pointer to a vector of pointers to hash_entrys, and the size of the
+   vector.  A hash_entry contains a union of all the info we like to
+   store in hash table.  If there is a hash collision, hash_entries
+   with the same hash are kept in a chain.  */
+
+/* What the data in a hash_entry means.  */
+typedef enum {
+  hash_integer,			/* Name->integer mapping.  */
+  hash_string,			/* Name->string mapping.  */
+  hash_macro,			/* Name is a macro.  */
+  hash_formal			/* Name is a formal argument.  */
+} hash_type;
+
+typedef struct hs {
+  sb key;			/* Symbol name.  */
+  hash_type type;		/* Symbol meaning.  */
+  union {
+    sb s;
+    int i;
+    struct macro_struct *m;
+    struct formal_struct *f;
+  } value;
+  struct hs *next;		/* Next hash_entry with same hash key.  */
+} hash_entry;
+
+typedef struct {
+  hash_entry **table;
+  int size;
+} hash_table;
+
+/* How we nest files and expand macros etc.
+
+   We keep a stack of of include_stack structs.  Each include file
+   pushes a new level onto the stack.  We keep an sb with a pushback
+   too.  unget chars are pushed onto the pushback sb, getchars first
+   checks the pushback sb before reading from the input stream.
+
+   Small things are expanded by adding the text of the item onto the
+   pushback sb.  Larger items are grown by pushing a new level and
+   allocating the entire pushback buf for the item.  Each time
+   something like a macro is expanded, the stack index is changed.  We
+   can then perform an exitm by popping all entries off the stack with
+   the same stack index.  If we're being reasonable, we can detect
+   recusive expansion by checking the index is reasonably small.  */
+
+typedef enum {
+  include_file, include_repeat, include_while, include_macro
+} include_type;
+
+struct include_stack {
+  sb pushback;			/* Current pushback stream.  */
+  int pushback_index;		/* Next char to read from stream.  */
+  FILE *handle;			/* Open file.  */
+  sb name;			/* Name of file.  */
+  int linecount;		/* Number of lines read so far.  */
+  include_type type;
+  int index;			/* Index of this layer.  */
+} include_stack[MAX_INCLUDES];
+
+struct include_stack *sp;
+#define isp (sp - include_stack)
+
+/* Include file list.  */
+
+typedef struct include_path {
+  struct include_path *next;
+  sb path;
+} include_path;
+
+include_path *paths_head;
+include_path *paths_tail;
+
+static void quit PARAMS ((void));
+static void hash_new_table PARAMS ((int, hash_table *));
+static int hash PARAMS ((sb *));
+static hash_entry *hash_create PARAMS ((hash_table *, sb *));
+static void hash_add_to_string_table PARAMS ((hash_table *, sb *, sb *, int));
+static void hash_add_to_int_table PARAMS ((hash_table *, sb *, int));
+static hash_entry *hash_lookup PARAMS ((hash_table *, sb *));
+static void checkconst PARAMS ((int, exp_t *));
+static int is_flonum PARAMS ((int, sb *));
+static int chew_flonum PARAMS ((int, sb *, sb *));
+static int sb_strtol PARAMS ((int, sb *, int, int *));
+static int level_0 PARAMS ((int, sb *, exp_t *));
+static int level_1 PARAMS ((int, sb *, exp_t *));
+static int level_2 PARAMS ((int, sb *, exp_t *));
+static int level_3 PARAMS ((int, sb *, exp_t *));
+static int level_4 PARAMS ((int, sb *, exp_t *));
+static int level_5 PARAMS ((int, sb *, exp_t *));
+static int exp_parse PARAMS ((int, sb *, exp_t *));
+static void exp_string PARAMS ((exp_t *, sb *));
+static int exp_get_abs PARAMS ((const char *, int, sb *, int *));
+#if 0
+static void strip_comments PARAMS ((sb *));
+#endif
+static void unget PARAMS ((int));
+static void include_buf PARAMS ((sb *, sb *, include_type, int));
+static void include_print_where_line PARAMS ((FILE *));
+static void include_print_line PARAMS ((FILE *));
+static int get_line PARAMS ((sb *));
+static int grab_label PARAMS ((sb *, sb *));
+static void change_base PARAMS ((int, sb *, sb *));
+static void do_end PARAMS ((sb *));
+static void do_assign PARAMS ((int, int, sb *));
+static void do_radix PARAMS ((sb *));
+static int get_opsize PARAMS ((int, sb *, int *));
+static int eol PARAMS ((int, sb *));
+static void do_data PARAMS ((int, sb *, int));
+static void do_datab PARAMS ((int, sb *));
+static void do_align PARAMS ((int, sb *));
+static void do_res PARAMS ((int, sb *, int));
+static void do_export PARAMS ((sb *));
+static void do_print PARAMS ((int, sb *));
+static void do_heading PARAMS ((int, sb *));
+static void do_page PARAMS ((void));
+static void do_form PARAMS ((int, sb *));
+static int get_any_string PARAMS ((int, sb *, sb *, int, int));
+static int skip_openp PARAMS ((int, sb *));
+static int skip_closep PARAMS ((int, sb *));
+static int dolen PARAMS ((int, sb *, sb *));
+static int doinstr PARAMS ((int, sb *, sb *));
+static int dosubstr PARAMS ((int, sb *, sb *));
+static void process_assigns PARAMS ((int, sb *, sb *));
+static int get_and_process PARAMS ((int, sb *, sb *));
+static void process_file PARAMS ((void));
+static void free_old_entry PARAMS ((hash_entry *));
+static void do_assigna PARAMS ((int, sb *));
+static void do_assignc PARAMS ((int, sb *));
+static void do_reg PARAMS ((int, sb *));
+static int condass_lookup_name PARAMS ((sb *, int, sb *, int));
+static int whatcond PARAMS ((int, sb *, int *));
+static int istrue PARAMS ((int, sb *));
+static void do_aif PARAMS ((int, sb *));
+static void do_aelse PARAMS ((void));
+static void do_aendi PARAMS ((void));
+static int condass_on PARAMS ((void));
+static void do_if PARAMS ((int, sb *, int));
+static int get_mri_string PARAMS ((int, sb *, sb *, int));
+static void do_ifc PARAMS ((int, sb *, int));
+static void do_aendr PARAMS ((void));
+static void do_awhile PARAMS ((int, sb *));
+static void do_aendw PARAMS ((void));
+static void do_exitm PARAMS ((void));
+static void do_arepeat PARAMS ((int, sb *));
+static void do_endm PARAMS ((void));
+static void do_irp PARAMS ((int, sb *, int));
+static void do_local PARAMS ((int, sb *));
+static void do_macro PARAMS ((int, sb *));
+static int macro_op PARAMS ((int, sb *));
+static int getstring PARAMS ((int, sb *, sb *));
+static void do_sdata PARAMS ((int, sb *, int));
+static void do_sdatab PARAMS ((int, sb *));
+static int new_file PARAMS ((const char *));
+static void do_include PARAMS ((int, sb *));
+static void include_pop PARAMS ((void));
+static int get PARAMS ((void));
+static int linecount PARAMS ((void));
+static int include_next_index PARAMS ((void));
+static void chartype_init PARAMS ((void));
+static int process_pseudo_op PARAMS ((int, sb *, sb *));
+static void add_keyword PARAMS ((const char *, int));
+static void process_init PARAMS ((void));
+static void do_define PARAMS ((const char *));
+static void show_usage PARAMS ((FILE *, int));
+static void show_help PARAMS ((void));
+
+/* --- functions recently removed from sb.c --- */
+
+/* put a null at the end of the sb at in and return the start of the
+ *    string, so that it can be used as an arg to printf %s.  */
+
+static char *
+sb_name (sb *in)
+{
+  /* stick a null on the end of the string */
+  sb_add_char (in, 0);
+  return in->ptr;
+}
+
+/* print the sb at ptr to the output file */
+
+static void
+sb_print (FILE *outfile, sb *ptr)
+{
+  int i;
+  int nc = 0;
+
+  for (i = 0; i < ptr->len; i++)
+    {
+      if (nc)
+	{
+	  fprintf (outfile, ",");
+	}
+      fprintf (outfile, "%d", ptr->ptr[i]);
+      nc = 1;
+    }
+}
+
+/* print the sb at ptr to the output file */
+
+static void
+sb_print_at (FILE *outfile, int idx, sb *ptr)
+{
+  int i;
+  for (i = idx; i < ptr->len; i++)
+    putc (ptr->ptr[i], outfile);
+}
+
+#define FATAL(x)				\
+  do						\
+    {						\
+      include_print_where_line (stderr);	\
+      fprintf x;				\
+      fatals++;					\
+      quit ();					\
+    }						\
+  while (0)
+
+#define ERROR(x)				\
+  do						\
+    {						\
+      include_print_where_line (stderr);	\
+      fprintf x;				\
+      errors++;					\
+    }						\
+  while (0)
+
+#define WARNING(x)				\
+  do						\
+    {						\
+      include_print_where_line (stderr);	\
+      fprintf x;				\
+      warnings++;				\
+    }						\
+  while (0)
+
+/* Exit the program and return the right ERROR code.  */
+
+static void
+quit ()
+{
+  int exitcode;
+  if (fatals + errors)
+    exitcode = 1;
+  else
+    exitcode = 0;
+
+  exit (exitcode);
+}
+
+/* Hash table maintenance.  */
+
+/* Build a new hash table with size buckets
+   and fill in the info at ptr.  */
+
+static void
+hash_new_table (size, ptr)
+     int size;
+     hash_table *ptr;
+{
+  int i;
+  ptr->size = size;
+  ptr->table = (hash_entry **) xmalloc (size * (sizeof (hash_entry *)));
+  /* Fill with null-pointer, not zero-bit-pattern.  */
+  for (i = 0; i < size; i++)
+    ptr->table[i] = 0;
+}
+
+/* Calculate and return the hash value of the sb at key.  */
+
+static int
+hash (key)
+     sb *key;
+{
+  int k = 0x1234;
+  int i;
+  char *p = key->ptr;
+  for (i = 0; i < key->len; i++)
+    {
+      k ^= (k << 2) ^ *p;
+      p++;
+    }
+  return k & 0xf0fff;
+}
+
+/* Look up key in hash_table tab.  If present, then return it,
+   otherwise build a new one and fill it with hash_integer.  */
+
+static hash_entry *
+hash_create (tab, key)
+     hash_table *tab;
+     sb *key;
+{
+  int k = hash (key) % tab->size;
+  hash_entry *p;
+  hash_entry **table = tab->table;
+
+  p = table[k];
+
+  while (1)
+    {
+      if (!p)
+	{
+	  hash_entry *n = (hash_entry *) xmalloc (sizeof (hash_entry));
+	  n->next = table[k];
+	  sb_new (&n->key);
+	  sb_add_sb (&n->key, key);
+	  table[k] = n;
+	  n->type = hash_integer;
+	  return n;
+	}
+      if (strncmp (table[k]->key.ptr, key->ptr, key->len) == 0)
+	{
+	  return p;
+	}
+      p = p->next;
+    }
+}
+
+/* Add sb name with key into hash_table tab.
+   If replacing old value and again, then ERROR.  */
+
+static void
+hash_add_to_string_table (tab, key, name, again)
+     hash_table *tab;
+     sb *key;
+     sb *name;
+     int again;
+{
+  hash_entry *ptr = hash_create (tab, key);
+  if (ptr->type == hash_integer)
+    {
+      sb_new (&ptr->value.s);
+    }
+  if (ptr->value.s.len)
+    {
+      if (!again)
+	ERROR ((stderr, _("redefinition not allowed\n")));
+    }
+
+  ptr->type = hash_string;
+  sb_reset (&ptr->value.s);
+
+  sb_add_sb (&ptr->value.s, name);
+}
+
+/* Add integer name to hash_table tab with sb key.  */
+
+static void
+hash_add_to_int_table (tab, key, name)
+     hash_table *tab;
+     sb *key;
+     int name;
+{
+  hash_entry *ptr = hash_create (tab, key);
+  ptr->value.i = name;
+}
+
+/* Look up sb key in hash_table tab.
+   If found, return hash_entry result, else 0.  */
+
+static hash_entry *
+hash_lookup (tab, key)
+     hash_table *tab;
+     sb *key;
+{
+  int k = hash (key) % tab->size;
+  hash_entry **table = tab->table;
+  hash_entry *p = table[k];
+  while (p)
+    {
+      if (p->key.len == key->len
+	  && strncmp (p->key.ptr, key->ptr, key->len) == 0)
+	return p;
+      p = p->next;
+    }
+  return 0;
+}
+
+/* expressions
+
+   are handled in a really simple recursive decent way. each bit of
+   the machine takes an index into an sb and a pointer to an exp_t,
+   modifies the *exp_t and returns the index of the first character
+   past the part of the expression parsed.
+
+ expression precedence:
+  ( )
+ unary + - ~
+  * /
+  + -
+  &
+  | ~
+*/
+
+/* Make sure that the exp_t at term is constant.
+   If not the give the op ERROR.  */
+
+static void
+checkconst (op, term)
+     int op;
+     exp_t *term;
+{
+  if (term->add_symbol.len
+      || term->sub_symbol.len)
+    {
+      ERROR ((stderr, _("the %c operator cannot take non-absolute arguments.\n"), op));
+    }
+}
+
+/* Chew the flonum from the string starting at idx.  Adjust idx to
+   point to the next character after the flonum.  */
+
+static int
+chew_flonum (idx, string, out)
+     int idx;
+     sb *string;
+     sb *out;
+{
+  sb buf;
+  regex_t reg;
+  regmatch_t match;
+
+  /* Duplicate and null terminate `string'.  */
+  sb_new (&buf);
+  sb_add_sb (&buf, string);
+  sb_add_char (&buf, '\0');
+
+  if (regcomp (&reg, "([0-9]*\\.[0-9]+([eE][+-]?[0-9]+)?)", REG_EXTENDED) != 0)
+    return idx;
+  if (regexec (&reg, &buf.ptr[idx], 1, &match, 0) != 0)
+    return idx;
+
+  /* Copy the match to the output.  */
+  assert (match.rm_eo >= match.rm_so);
+  sb_add_buffer (out, &buf.ptr[idx], match.rm_eo - match.rm_so);
+
+  sb_kill (&buf);
+  regfree (&reg);
+  idx += match.rm_eo;
+  return idx;
+}
+
+static int
+is_flonum (idx, string)
+     int idx;
+     sb *string;
+{
+  sb buf;
+  regex_t reg;
+  int rc;
+
+  /* Duplicate and null terminate `string'.  */
+  sb_new (&buf);
+  sb_add_sb (&buf, string);
+  sb_add_char (&buf, '\0');
+
+  if (regcomp (&reg, "^[0-9]*\\.[0-9]+([eE][+-]?[0-9]+)?", REG_EXTENDED) != 0)
+    return 0;
+
+  rc = regexec (&reg, &buf.ptr[idx], 0, NULL, 0);
+  sb_kill (&buf);
+  regfree (&reg);
+  return (rc == 0);
+}
+
+/* Turn the number in string at idx into a number of base, fill in
+   ptr, and return the index of the first character not in the number.  */
+
+static int
+sb_strtol (idx, string, base, ptr)
+     int idx;
+     sb *string;
+     int base;
+     int *ptr;
+{
+  int value = 0;
+  idx = sb_skip_white (idx, string);
+
+  while (idx < string->len)
+    {
+      int ch = string->ptr[idx];
+      int dig = 0;
+      if (ISDIGIT (ch))
+	dig = ch - '0';
+      else if (ch >= 'a' && ch <= 'f')
+	dig = ch - 'a' + 10;
+      else if (ch >= 'A' && ch <= 'F')
+	dig = ch - 'A' + 10;
+      else
+	break;
+
+      if (dig >= base)
+	break;
+
+      value = value * base + dig;
+      idx++;
+    }
+  *ptr = value;
+  return idx;
+}
+
+static int
+level_0 (idx, string, lhs)
+     int idx;
+     sb *string;
+     exp_t *lhs;
+{
+  lhs->add_symbol.len = 0;
+  lhs->add_symbol.name = 0;
+
+  lhs->sub_symbol.len = 0;
+  lhs->sub_symbol.name = 0;
+
+  idx = sb_skip_white (idx, string);
+
+  lhs->value = 0;
+
+  if (ISDIGIT (string->ptr[idx]))
+    {
+      idx = sb_strtol (idx, string, 10, &lhs->value);
+    }
+  else if (ISFIRSTCHAR (string->ptr[idx]))
+    {
+      int len = 0;
+      lhs->add_symbol.name = string->ptr + idx;
+      while (idx < string->len && ISNEXTCHAR (string->ptr[idx]))
+	{
+	  idx++;
+	  len++;
+	}
+      lhs->add_symbol.len = len;
+    }
+  else if (string->ptr[idx] == '"')
+    {
+      sb acc;
+      sb_new (&acc);
+      ERROR ((stderr, _("string where expression expected.\n")));
+      idx = getstring (idx, string, &acc);
+      sb_kill (&acc);
+    }
+  else
+    {
+      ERROR ((stderr, _("can't find primary in expression.\n")));
+      idx++;
+    }
+  return sb_skip_white (idx, string);
+}
+
+static int
+level_1 (idx, string, lhs)
+     int idx;
+     sb *string;
+     exp_t *lhs;
+{
+  idx = sb_skip_white (idx, string);
+
+  switch (string->ptr[idx])
+    {
+    case '+':
+      idx = level_1 (idx + 1, string, lhs);
+      break;
+    case '~':
+      idx = level_1 (idx + 1, string, lhs);
+      checkconst ('~', lhs);
+      lhs->value = ~lhs->value;
+      break;
+    case '-':
+      {
+	symbol t;
+	idx = level_1 (idx + 1, string, lhs);
+	lhs->value = -lhs->value;
+	t = lhs->add_symbol;
+	lhs->add_symbol = lhs->sub_symbol;
+	lhs->sub_symbol = t;
+	break;
+      }
+    case '(':
+      idx++;
+      idx = level_5 (sb_skip_white (idx, string), string, lhs);
+      if (string->ptr[idx] != ')')
+	ERROR ((stderr, _("misplaced closing parens.\n")));
+      else
+	idx++;
+      break;
+    default:
+      idx = level_0 (idx, string, lhs);
+      break;
+    }
+  return sb_skip_white (idx, string);
+}
+
+static int
+level_2 (idx, string, lhs)
+     int idx;
+     sb *string;
+     exp_t *lhs;
+{
+  exp_t rhs;
+
+  idx = level_1 (idx, string, lhs);
+
+  while (idx < string->len && (string->ptr[idx] == '*'
+			       || string->ptr[idx] == '/'))
+    {
+      char op = string->ptr[idx++];
+      idx = level_1 (idx, string, &rhs);
+      switch (op)
+	{
+	case '*':
+	  checkconst ('*', lhs);
+	  checkconst ('*', &rhs);
+	  lhs->value *= rhs.value;
+	  break;
+	case '/':
+	  checkconst ('/', lhs);
+	  checkconst ('/', &rhs);
+	  if (rhs.value == 0)
+	    ERROR ((stderr, _("attempt to divide by zero.\n")));
+	  else
+	    lhs->value /= rhs.value;
+	  break;
+	}
+    }
+  return sb_skip_white (idx, string);
+}
+
+static int
+level_3 (idx, string, lhs)
+     int idx;
+     sb *string;
+     exp_t *lhs;
+{
+  exp_t rhs;
+
+  idx = level_2 (idx, string, lhs);
+
+  while (idx < string->len
+	 && (string->ptr[idx] == '+'
+	     || string->ptr[idx] == '-'))
+    {
+      char op = string->ptr[idx++];
+      idx = level_2 (idx, string, &rhs);
+      switch (op)
+	{
+	case '+':
+	  lhs->value += rhs.value;
+	  if (lhs->add_symbol.name && rhs.add_symbol.name)
+	    {
+	      ERROR ((stderr, _("can't add two relocatable expressions\n")));
+	    }
+	  /* Change nn+symbol to symbol + nn.  */
+	  if (rhs.add_symbol.name)
+	    {
+	      lhs->add_symbol = rhs.add_symbol;
+	    }
+	  break;
+	case '-':
+	  lhs->value -= rhs.value;
+	  lhs->sub_symbol = rhs.add_symbol;
+	  break;
+	}
+    }
+  return sb_skip_white (idx, string);
+}
+
+static int
+level_4 (idx, string, lhs)
+     int idx;
+     sb *string;
+     exp_t *lhs;
+{
+  exp_t rhs;
+
+  idx = level_3 (idx, string, lhs);
+
+  while (idx < string->len &&
+	 string->ptr[idx] == '&')
+    {
+      char op = string->ptr[idx++];
+      idx = level_3 (idx, string, &rhs);
+      switch (op)
+	{
+	case '&':
+	  checkconst ('&', lhs);
+	  checkconst ('&', &rhs);
+	  lhs->value &= rhs.value;
+	  break;
+	}
+    }
+  return sb_skip_white (idx, string);
+}
+
+static int
+level_5 (idx, string, lhs)
+     int idx;
+     sb *string;
+     exp_t *lhs;
+{
+  exp_t rhs;
+
+  idx = level_4 (idx, string, lhs);
+
+  while (idx < string->len
+	 && (string->ptr[idx] == '|' || string->ptr[idx] == '~'))
+    {
+      char op = string->ptr[idx++];
+      idx = level_4 (idx, string, &rhs);
+      switch (op)
+	{
+	case '|':
+	  checkconst ('|', lhs);
+	  checkconst ('|', &rhs);
+	  lhs->value |= rhs.value;
+	  break;
+	case '~':
+	  checkconst ('~', lhs);
+	  checkconst ('~', &rhs);
+	  lhs->value ^= rhs.value;
+	  break;
+	}
+    }
+  return sb_skip_white (idx, string);
+}
+
+/* Parse the expression at offset idx into string, fill up res with
+   the result.  Return the index of the first char past the
+   expression.  */
+
+static int
+exp_parse (idx, string, res)
+     int idx;
+     sb *string;
+     exp_t *res;
+{
+  return level_5 (sb_skip_white (idx, string), string, res);
+}
+
+/* Turn the expression at exp into text and glue it onto the end of
+   string.  */
+
+static void
+exp_string (exp, string)
+     exp_t *exp;
+     sb *string;
+{
+  int np = 0;
+  int ad = 0;
+  sb_reset (string);
+
+  if (exp->add_symbol.len)
+    {
+      sb_add_buffer (string, exp->add_symbol.name, exp->add_symbol.len);
+      np = 1;
+      ad = 1;
+    }
+  if (exp->value)
+    {
+      char buf[20];
+      if (np)
+	sb_add_char (string, '+');
+      sprintf (buf, "%d", exp->value);
+      sb_add_string (string, buf);
+      np = 1;
+      ad = 1;
+    }
+  if (exp->sub_symbol.len)
+    {
+      sb_add_char (string, '-');
+      sb_add_buffer (string, exp->add_symbol.name, exp->add_symbol.len);
+      np = 0;
+      ad = 1;
+    }
+
+  if (!ad)
+    sb_add_char (string, '0');
+}
+
+/* Parse the expression at offset idx into sb in.  Return the value in
+   val.  If the expression is not constant, give ERROR emsg.  Return
+   the index of the first character past the end of the expression.  */
+
+static int
+exp_get_abs (emsg, idx, in, val)
+     const char *emsg;
+     int idx;
+     sb *in;
+     int *val;
+{
+  exp_t res;
+  idx = exp_parse (idx, in, &res);
+  if (res.add_symbol.len || res.sub_symbol.len)
+    ERROR ((stderr, "%s", emsg));
+  *val = res.value;
+  return idx;
+}
+
+/* Current label parsed from line.  */
+sb label;
+
+/* Hash table for all assigned variables.  */
+hash_table assign_hash_table;
+
+/* Hash table for keyword.  */
+hash_table keyword_hash_table;
+
+/* Hash table for eq variables.  */
+hash_table vars;
+
+#define in_comment ';'
+
+#if 0
+static void
+strip_comments (out)
+     sb *out;
+{
+  char *s = out->ptr;
+  int i = 0;
+  for (i = 0; i < out->len; i++)
+    {
+      if (ISCOMMENTCHAR (s[i]))
+	{
+	  out->len = i;
+	  return;
+	}
+    }
+}
+#endif
+
+/* Push back character ch so that it can be read again.  */
+
+static void
+unget (ch)
+     int ch;
+{
+  if (ch == '\n')
+    {
+      sp->linecount--;
+    }
+  if (sp->pushback_index)
+    sp->pushback_index--;
+  else
+    sb_add_char (&sp->pushback, ch);
+}
+
+/* Push the sb ptr onto the include stack, with the given name, type
+   and index.  */
+
+static void
+include_buf (name, ptr, type, index)
+     sb *name;
+     sb *ptr;
+     include_type type;
+     int index;
+{
+  sp++;
+  if (sp - include_stack >= MAX_INCLUDES)
+    FATAL ((stderr, _("unreasonable nesting.\n")));
+  sb_new (&sp->name);
+  sb_add_sb (&sp->name, name);
+  sp->handle = 0;
+  sp->linecount = 1;
+  sp->pushback_index = 0;
+  sp->type = type;
+  sp->index = index;
+  sb_new (&sp->pushback);
+  sb_add_sb (&sp->pushback, ptr);
+}
+
+/* Used in ERROR messages, print info on where the include stack is
+   onto file.  */
+
+static void
+include_print_where_line (file)
+     FILE *file;
+{
+  struct include_stack *p = include_stack + 1;
+
+  while (p <= sp)
+    {
+      fprintf (file, "%s:%d ", sb_name (&p->name), p->linecount - 1);
+      p++;
+    }
+}
+
+/* Used in listings, print the line number onto file.  */
+
+static void
+include_print_line (file)
+     FILE *file;
+{
+  int n;
+  struct include_stack *p = include_stack + 1;
+
+  n = fprintf (file, "%4d", p->linecount);
+  p++;
+  while (p <= sp)
+    {
+      n += fprintf (file, ".%d", p->linecount);
+      p++;
+    }
+  while (n < 8 * 3)
+    {
+      fprintf (file, " ");
+      n++;
+    }
+}
+
+/* Read a line from the top of the include stack into sb in.  */
+
+static int
+get_line (in)
+     sb *in;
+{
+  int online = 0;
+  int more = 1, ch = 0;
+
+  if (copysource)
+    {
+      putc (comment_char, outfile);
+      if (print_line_number)
+	include_print_line (outfile);
+    }
+
+  while (1)
+    {
+      ch = get ();
+
+      while (ch == '\r')
+	ch = get ();
+
+      if (ch == EOF)
+	{
+	  if (online)
+	    {
+	      WARNING ((stderr, _("End of file not at start of line.\n")));
+	      if (copysource)
+		putc ('\n', outfile);
+	      ch = '\n';
+	    }
+	  else
+	    more = 0;
+	  break;
+	}
+
+      if (copysource)
+	{
+	  putc (ch, outfile);
+	}
+
+      if (ch == '\n')
+	{
+	  ch = get ();
+	  online = 0;
+	  if (ch == '+')
+	    {
+	      /* Continued line.  */
+	      if (copysource)
+		{
+		  putc (comment_char, outfile);
+		  putc ('+', outfile);
+		}
+	      ch = get ();
+	    }
+	  else
+	    {
+	      if (ch != EOF)
+		unget (ch);
+	      ch = '\n'; break;
+	    }
+	}
+      else
+	{
+	  sb_add_char (in, ch);
+	}
+      online++;
+    }
+
+  return more ? ch : 0;
+}
+
+/* Find a label from sb in and put it in out.  */
+
+static int
+grab_label (in, out)
+     sb *in;
+     sb *out;
+{
+  int i = 0;
+  sb_reset (out);
+  if (ISFIRSTCHAR (in->ptr[i]) || in->ptr[i] == '\\')
+    {
+      sb_add_char (out, in->ptr[i]);
+      i++;
+      while ((ISNEXTCHAR (in->ptr[i])
+	      || in->ptr[i] == '\\'
+	      || in->ptr[i] == '&')
+	     && i < in->len)
+	{
+	  sb_add_char (out, in->ptr[i]);
+	  i++;
+	}
+    }
+  return i;
+}
+
+/* Find all strange base stuff and turn into decimal.  Also
+   find all the other numbers and convert them from the default radix.  */
+
+static void
+change_base (idx, in, out)
+     int idx;
+     sb *in;
+     sb *out;
+{
+  char buffer[20];
+
+  while (idx < in->len)
+    {
+      if (in->ptr[idx] == '\\'
+	  && idx + 1 < in->len
+	  && in->ptr[idx + 1] == '(')
+	{
+	  idx += 2;
+	  while (idx < in->len
+		 && in->ptr[idx] != ')')
+	    {
+	      sb_add_char (out, in->ptr[idx]);
+	      idx++;
+	    }
+	  if (idx < in->len)
+	    idx++;
+	}
+      else if (idx < in->len - 1 && in->ptr[idx + 1] == '\'' && ! mri)
+	{
+	  int base;
+	  int value;
+	  switch (in->ptr[idx])
+	    {
+	    case 'b':
+	    case 'B':
+	      base = 2;
+	      break;
+	    case 'q':
+	    case 'Q':
+	      base = 8;
+	      break;
+	    case 'h':
+	    case 'H':
+	      base = 16;
+	      break;
+	    case 'd':
+	    case 'D':
+	      base = 10;
+	      break;
+	    default:
+	      ERROR ((stderr, _("Illegal base character %c.\n"), in->ptr[idx]));
+	      base = 10;
+	      break;
+	    }
+
+	  idx = sb_strtol (idx + 2, in, base, &value);
+	  sprintf (buffer, "%d", value);
+	  sb_add_string (out, buffer);
+	}
+      else if (ISFIRSTCHAR (in->ptr[idx]))
+	{
+	  /* Copy entire names through quickly.  */
+	  sb_add_char (out, in->ptr[idx]);
+	  idx++;
+	  while (idx < in->len && ISNEXTCHAR (in->ptr[idx]))
+	    {
+	      sb_add_char (out, in->ptr[idx]);
+	      idx++;
+	    }
+	}
+      else if (is_flonum (idx, in))
+	{
+	  idx = chew_flonum (idx, in, out);
+	}
+      else if (ISDIGIT (in->ptr[idx]))
+	{
+	  int value;
+	  /* All numbers must start with a digit, let's chew it and
+	     spit out decimal.  */
+	  idx = sb_strtol (idx, in, radix, &value);
+	  sprintf (buffer, "%d", value);
+	  sb_add_string (out, buffer);
+
+	  /* Skip all undigsested letters.  */
+	  while (idx < in->len && ISNEXTCHAR (in->ptr[idx]))
+	    {
+	      sb_add_char (out, in->ptr[idx]);
+	      idx++;
+	    }
+	}
+      else if (in->ptr[idx] == '"' || in->ptr[idx] == '\'')
+	{
+	  char tchar = in->ptr[idx];
+	  /* Copy entire names through quickly.  */
+	  sb_add_char (out, in->ptr[idx]);
+	  idx++;
+	  while (idx < in->len && in->ptr[idx] != tchar)
+	    {
+	      sb_add_char (out, in->ptr[idx]);
+	      idx++;
+	    }
+	}
+      else
+	{
+	  /* Nothing special, just pass it through.  */
+	  sb_add_char (out, in->ptr[idx]);
+	  idx++;
+	}
+    }
+
+}
+
+/* .end  */
+
+static void
+do_end (in)
+     sb *in;
+{
+  had_end = 1;
+  if (mri)
+    fprintf (outfile, "%s\n", sb_name (in));
+}
+
+/* .assign  */
+
+static void
+do_assign (again, idx, in)
+     int again;
+     int idx;
+     sb *in;
+{
+  /* Stick label in symbol table with following value.  */
+  exp_t e;
+  sb acc;
+
+  sb_new (&acc);
+  idx = exp_parse (idx, in, &e);
+  exp_string (&e, &acc);
+  hash_add_to_string_table (&assign_hash_table, &label, &acc, again);
+  sb_kill (&acc);
+}
+
+/* .radix [b|q|d|h]  */
+
+static void
+do_radix (ptr)
+     sb *ptr;
+{
+  int idx = sb_skip_white (0, ptr);
+  switch (ptr->ptr[idx])
+    {
+    case 'B':
+    case 'b':
+      radix = 2;
+      break;
+    case 'q':
+    case 'Q':
+      radix = 8;
+      break;
+    case 'd':
+    case 'D':
+      radix = 10;
+      break;
+    case 'h':
+    case 'H':
+      radix = 16;
+      break;
+    default:
+      ERROR ((stderr, _("radix is %c must be one of b, q, d or h"), radix));
+    }
+}
+
+/* Parse off a .b, .w or .l.  */
+
+static int
+get_opsize (idx, in, size)
+     int idx;
+     sb *in;
+     int *size;
+{
+  *size = 4;
+  if (in->ptr[idx] == '.')
+    {
+      idx++;
+    }
+  switch (in->ptr[idx])
+    {
+    case 'b':
+    case 'B':
+      *size = 1;
+      break;
+    case 'w':
+    case 'W':
+      *size = 2;
+      break;
+    case 'l':
+    case 'L':
+      *size = 4;
+      break;
+    case ' ':
+    case '\t':
+      break;
+    default:
+      ERROR ((stderr, _("size must be one of b, w or l, is %c.\n"), in->ptr[idx]));
+      break;
+    }
+  idx++;
+
+  return idx;
+}
+
+static int
+eol (idx, line)
+     int idx;
+     sb *line;
+{
+  idx = sb_skip_white (idx, line);
+  if (idx < line->len
+      && ISCOMMENTCHAR(line->ptr[idx]))
+    return 1;
+  if (idx >= line->len)
+    return 1;
+  return 0;
+}
+
+/* .data [.b|.w|.l] <data>*
+    or d[bwl] <data>*  */
+
+static void
+do_data (idx, in, size)
+     int idx;
+     sb *in;
+     int size;
+{
+  int opsize = 4;
+  char *opname = ".yikes!";
+  sb acc;
+  sb_new (&acc);
+
+  if (!size)
+    {
+      idx = get_opsize (idx, in, &opsize);
+    }
+  else
+    {
+      opsize = size;
+    }
+  switch (opsize)
+    {
+    case 4:
+      opname = ".long";
+      break;
+    case 2:
+      opname = ".short";
+      break;
+    case 1:
+      opname = ".byte";
+      break;
+    }
+
+  fprintf (outfile, "%s\t", opname);
+
+  idx = sb_skip_white (idx, in);
+
+  if (alternate
+      && idx < in->len
+      && in->ptr[idx] == '"')
+    {
+      int i;
+      idx = getstring (idx, in, &acc);
+      for (i = 0; i < acc.len; i++)
+	{
+	  if (i)
+	    fprintf (outfile, ",");
+	  fprintf (outfile, "%d", acc.ptr[i]);
+	}
+    }
+  else
+    {
+      while (!eol (idx, in))
+	{
+	  exp_t e;
+	  idx = exp_parse (idx, in, &e);
+	  exp_string (&e, &acc);
+	  sb_add_char (&acc, 0);
+	  fprintf (outfile, "%s", acc.ptr);
+	  if (idx < in->len && in->ptr[idx] == ',')
+	    {
+	      fprintf (outfile, ",");
+	      idx++;
+	    }
+	}
+    }
+  sb_kill (&acc);
+  sb_print_at (outfile, idx, in);
+  fprintf (outfile, "\n");
+}
+
+/* .datab [.b|.w|.l] <repeat>,<fill>  */
+
+static void
+do_datab (idx, in)
+     int idx;
+     sb *in;
+{
+  int opsize;
+  int repeat;
+  int fill;
+
+  idx = get_opsize (idx, in, &opsize);
+
+  idx = exp_get_abs (_("datab repeat must be constant.\n"), idx, in, &repeat);
+  idx = sb_skip_comma (idx, in);
+  idx = exp_get_abs (_("datab data must be absolute.\n"), idx, in, &fill);
+
+  fprintf (outfile, ".fill\t%d,%d,%d\n", repeat, opsize, fill);
+}
+
+/* .align <size>  */
+
+static void
+do_align (idx, in)
+     int idx;
+     sb *in;
+{
+  int al, have_fill, fill;
+
+  idx = exp_get_abs (_("align needs absolute expression.\n"), idx, in, &al);
+  idx = sb_skip_white (idx, in);
+  have_fill = 0;
+  fill = 0;
+  if (! eol (idx, in))
+    {
+      idx = sb_skip_comma (idx, in);
+      idx = exp_get_abs (_(".align needs absolute fill value.\n"), idx, in,
+			 &fill);
+      have_fill = 1;
+    }
+
+  fprintf (outfile, ".align	%d", al);
+  if (have_fill)
+    fprintf (outfile, ",%d", fill);
+  fprintf (outfile, "\n");
+}
+
+/* .res[.b|.w|.l] <size>  */
+
+static void
+do_res (idx, in, type)
+     int idx;
+     sb *in;
+     int type;
+{
+  int size = 4;
+  int count = 0;
+
+  idx = get_opsize (idx, in, &size);
+  while (!eol (idx, in))
+    {
+      idx = sb_skip_white (idx, in);
+      if (in->ptr[idx] == ',')
+	idx++;
+      idx = exp_get_abs (_("res needs absolute expression for fill count.\n"), idx, in, &count);
+
+      if (type == 'c' || type == 'z')
+	count++;
+
+      fprintf (outfile, ".space	%d\n", count * size);
+    }
+}
+
+/* .export  */
+
+static void
+do_export (in)
+     sb *in;
+{
+  fprintf (outfile, ".global	%s\n", sb_name (in));
+}
+
+/* .print [list] [nolist]  */
+
+static void
+do_print (idx, in)
+     int idx;
+     sb *in;
+{
+  idx = sb_skip_white (idx, in);
+  while (idx < in->len)
+    {
+      if (strncasecmp (in->ptr + idx, "LIST", 4) == 0)
+	{
+	  fprintf (outfile, ".list\n");
+	  idx += 4;
+	}
+      else if (strncasecmp (in->ptr + idx, "NOLIST", 6) == 0)
+	{
+	  fprintf (outfile, ".nolist\n");
+	  idx += 6;
+	}
+      idx++;
+    }
+}
+
+/* .head  */
+
+static void
+do_heading (idx, in)
+     int idx;
+     sb *in;
+{
+  sb head;
+  sb_new (&head);
+  idx = getstring (idx, in, &head);
+  fprintf (outfile, ".title	\"%s\"\n", sb_name (&head));
+  sb_kill (&head);
+}
+
+/* .page  */
+
+static void
+do_page ()
+{
+  fprintf (outfile, ".eject\n");
+}
+
+/* .form [lin=<value>] [col=<value>]  */
+
+static void
+do_form (idx, in)
+     int idx;
+     sb *in;
+{
+  int lines = 60;
+  int columns = 132;
+  idx = sb_skip_white (idx, in);
+
+  while (idx < in->len)
+    {
+
+      if (strncasecmp (in->ptr + idx, "LIN=", 4) == 0)
+	{
+	  idx += 4;
+	  idx = exp_get_abs (_("form LIN= needs absolute expresssion.\n"), idx, in, &lines);
+	}
+
+      if (strncasecmp (in->ptr + idx, _("COL="), 4) == 0)
+	{
+	  idx += 4;
+	  idx = exp_get_abs (_("form COL= needs absolute expresssion.\n"), idx, in, &columns);
+	}
+
+      idx++;
+    }
+  fprintf (outfile, ".psize %d,%d\n", lines, columns);
+
+}
+
+/* Fetch string from the input stream,
+   rules:
+    'Bxyx<whitespace>  	-> return 'Bxyza
+    %<char>		-> return string of decimal value of x
+    "<string>"		-> return string
+    xyx<whitespace>     -> return xyz
+*/
+
+static int
+get_any_string (idx, in, out, expand, pretend_quoted)
+     int idx;
+     sb *in;
+     sb *out;
+     int expand;
+     int pretend_quoted;
+{
+  sb_reset (out);
+  idx = sb_skip_white (idx, in);
+
+  if (idx < in->len)
+    {
+      if (in->len > 2 && in->ptr[idx + 1] == '\'' && ISBASE (in->ptr[idx]))
+	{
+	  while (!ISSEP (in->ptr[idx]))
+	    sb_add_char (out, in->ptr[idx++]);
+	}
+      else if (in->ptr[idx] == '%'
+	       && alternate
+	       && expand)
+	{
+	  int val;
+	  char buf[20];
+	  /* Turns the next expression into a string.  */
+	  /* xgettext: no-c-format */
+	  idx = exp_get_abs (_("% operator needs absolute expression"),
+			     idx + 1,
+			     in,
+			     &val);
+	  sprintf (buf, "%d", val);
+	  sb_add_string (out, buf);
+	}
+      else if (in->ptr[idx] == '"'
+	       || in->ptr[idx] == '<'
+	       || (alternate && in->ptr[idx] == '\''))
+	{
+	  if (alternate && expand)
+	    {
+	      /* Keep the quotes.  */
+	      sb_add_char (out, '\"');
+
+	      idx = getstring (idx, in, out);
+	      sb_add_char (out, '\"');
+
+	    }
+	  else
+	    {
+	      idx = getstring (idx, in, out);
+	    }
+	}
+      else
+	{
+	  while (idx < in->len
+		 && (in->ptr[idx] == '"'
+		     || in->ptr[idx] == '\''
+		     || pretend_quoted
+		     || !ISSEP (in->ptr[idx])))
+	    {
+	      if (in->ptr[idx] == '"'
+		  || in->ptr[idx] == '\'')
+		{
+		  char tchar = in->ptr[idx];
+		  sb_add_char (out, in->ptr[idx++]);
+		  while (idx < in->len
+			 && in->ptr[idx] != tchar)
+		    sb_add_char (out, in->ptr[idx++]);
+		  if (idx == in->len)
+		    return idx;
+		}
+	      sb_add_char (out, in->ptr[idx++]);
+	    }
+	}
+    }
+
+  return idx;
+}
+
+/* Skip along sb in starting at idx, suck off whitespace a ( and more
+   whitespace.  Return the idx of the next char.  */
+
+static int
+skip_openp (idx, in)
+     int idx;
+     sb *in;
+{
+  idx = sb_skip_white (idx, in);
+  if (in->ptr[idx] != '(')
+    ERROR ((stderr, _("misplaced ( .\n")));
+  idx = sb_skip_white (idx + 1, in);
+  return idx;
+}
+
+/* Skip along sb in starting at idx, suck off whitespace a ) and more
+   whitespace.  Return the idx of the next char.  */
+
+static int
+skip_closep (idx, in)
+     int idx;
+     sb *in;
+{
+  idx = sb_skip_white (idx, in);
+  if (in->ptr[idx] != ')')
+    ERROR ((stderr, _("misplaced ).\n")));
+  idx = sb_skip_white (idx + 1, in);
+  return idx;
+}
+
+/* .len  */
+
+static int
+dolen (idx, in, out)
+     int idx;
+     sb *in;
+     sb *out;
+{
+
+  sb stringout;
+  char buffer[10];
+
+  sb_new (&stringout);
+  idx = skip_openp (idx, in);
+  idx = get_and_process (idx, in, &stringout);
+  idx = skip_closep (idx, in);
+  sprintf (buffer, "%d", stringout.len);
+  sb_add_string (out, buffer);
+
+  sb_kill (&stringout);
+  return idx;
+}
+
+/* .instr  */
+
+static int
+doinstr (idx, in, out)
+     int idx;
+     sb *in;
+     sb *out;
+{
+  sb string;
+  sb search;
+  int i;
+  int start;
+  int res;
+  char buffer[10];
+
+  sb_new (&string);
+  sb_new (&search);
+  idx = skip_openp (idx, in);
+  idx = get_and_process (idx, in, &string);
+  idx = sb_skip_comma (idx, in);
+  idx = get_and_process (idx, in, &search);
+  idx = sb_skip_comma (idx, in);
+  if (ISDIGIT (in->ptr[idx]))
+    {
+      idx = exp_get_abs (_(".instr needs absolute expresson.\n"), idx, in, &start);
+    }
+  else
+    {
+      start = 0;
+    }
+  idx = skip_closep (idx, in);
+  res = -1;
+  for (i = start; i < string.len; i++)
+    {
+      if (strncmp (string.ptr + i, search.ptr, search.len) == 0)
+	{
+	  res = i;
+	  break;
+	}
+    }
+  sprintf (buffer, "%d", res);
+  sb_add_string (out, buffer);
+  sb_kill (&string);
+  sb_kill (&search);
+  return idx;
+}
+
+static int
+dosubstr (idx, in, out)
+     int idx;
+     sb *in;
+     sb *out;
+{
+  sb string;
+  int pos;
+  int len;
+  sb_new (&string);
+
+  idx = skip_openp (idx, in);
+  idx = get_and_process (idx, in, &string);
+  idx = sb_skip_comma (idx, in);
+  idx = exp_get_abs (_("need absolute position.\n"), idx, in, &pos);
+  idx = sb_skip_comma (idx, in);
+  idx = exp_get_abs (_("need absolute length.\n"), idx, in, &len);
+  idx = skip_closep (idx, in);
+
+  if (len < 0 || pos < 0 ||
+      pos > string.len
+      || pos + len > string.len)
+    {
+      sb_add_string (out, " ");
+    }
+  else
+    {
+      sb_add_char (out, '"');
+      while (len > 0)
+	{
+	  sb_add_char (out, string.ptr[pos++]);
+	  len--;
+	}
+      sb_add_char (out, '"');
+    }
+  sb_kill (&string);
+  return idx;
+}
+
+/* Scan line, change tokens in the hash table to their replacements.  */
+
+static void
+process_assigns (idx, in, buf)
+     int idx;
+     sb *in;
+     sb *buf;
+{
+  while (idx < in->len)
+    {
+      hash_entry *ptr;
+      if (in->ptr[idx] == '\\'
+	  && idx + 1 < in->len
+	  && in->ptr[idx + 1] == '(')
+	{
+	  do
+	    {
+	      sb_add_char (buf, in->ptr[idx]);
+	      idx++;
+	    }
+	  while (idx < in->len && in->ptr[idx - 1] != ')');
+	}
+      else if (in->ptr[idx] == '\\'
+	  && idx + 1 < in->len
+	  && in->ptr[idx + 1] == '&')
+	{
+	  idx = condass_lookup_name (in, idx + 2, buf, 1);
+	}
+      else if (in->ptr[idx] == '\\'
+	       && idx + 1 < in->len
+	       && in->ptr[idx + 1] == '$')
+	{
+	  idx = condass_lookup_name (in, idx + 2, buf, 0);
+	}
+      else if (idx + 3 < in->len
+	       && in->ptr[idx] == '.'
+	       && TOUPPER (in->ptr[idx + 1]) == 'L'
+	       && TOUPPER (in->ptr[idx + 2]) == 'E'
+	       && TOUPPER (in->ptr[idx + 3]) == 'N')
+	idx = dolen (idx + 4, in, buf);
+      else if (idx + 6 < in->len
+	       && in->ptr[idx] == '.'
+	       && TOUPPER (in->ptr[idx + 1]) == 'I'
+	       && TOUPPER (in->ptr[idx + 2]) == 'N'
+	       && TOUPPER (in->ptr[idx + 3]) == 'S'
+	       && TOUPPER (in->ptr[idx + 4]) == 'T'
+	       && TOUPPER (in->ptr[idx + 5]) == 'R')
+	idx = doinstr (idx + 6, in, buf);
+      else if (idx + 7 < in->len
+	       && in->ptr[idx] == '.'
+	       && TOUPPER (in->ptr[idx + 1]) == 'S'
+	       && TOUPPER (in->ptr[idx + 2]) == 'U'
+	       && TOUPPER (in->ptr[idx + 3]) == 'B'
+	       && TOUPPER (in->ptr[idx + 4]) == 'S'
+	       && TOUPPER (in->ptr[idx + 5]) == 'T'
+	       && TOUPPER (in->ptr[idx + 6]) == 'R')
+	idx = dosubstr (idx + 7, in, buf);
+      else if (ISFIRSTCHAR (in->ptr[idx]))
+	{
+	  /* May be a simple name subsitution, see if we have a word.  */
+	  sb acc;
+	  int cur = idx + 1;
+	  while (cur < in->len
+		 && (ISNEXTCHAR (in->ptr[cur])))
+	    cur++;
+
+	  sb_new (&acc);
+	  sb_add_buffer (&acc, in->ptr + idx, cur - idx);
+	  ptr = hash_lookup (&assign_hash_table, &acc);
+	  if (ptr)
+	    {
+	      /* Found a definition for it.  */
+	      sb_add_sb (buf, &ptr->value.s);
+	    }
+	  else
+	    {
+	      /* No definition, just copy the word.  */
+	      sb_add_sb (buf, &acc);
+	    }
+	  sb_kill (&acc);
+	  idx = cur;
+	}
+      else
+	{
+	  sb_add_char (buf, in->ptr[idx++]);
+	}
+    }
+}
+
+static int
+get_and_process (idx, in, out)
+     int idx;
+     sb *in;
+     sb *out;
+{
+  sb t;
+  sb_new (&t);
+  idx = get_any_string (idx, in, &t, 1, 0);
+  process_assigns (0, &t, out);
+  sb_kill (&t);
+  return idx;
+}
+
+static void
+process_file ()
+{
+  sb line;
+  sb t1, t2;
+  sb acc;
+  sb label_in;
+  int more;
+
+  sb_new (&line);
+  sb_new (&t1);
+  sb_new (&t2);
+  sb_new (&acc);
+  sb_new (&label_in);
+  sb_reset (&line);
+  more = get_line (&line);
+  while (more)
+    {
+      /* Find any label and pseudo op that we're intested in.  */
+      int l;
+      if (line.len == 0)
+	{
+	  if (condass_on ())
+	    fprintf (outfile, "\n");
+	}
+      else if (mri
+	       && (line.ptr[0] == '*'
+		   || line.ptr[0] == '!'))
+	{
+	  /* MRI line comment.  */
+	  fprintf (outfile, "%s", sb_name (&line));
+	}
+      else
+	{
+	  l = grab_label (&line, &label_in);
+	  sb_reset (&label);
+
+	  if (line.ptr[l] == ':')
+	    l++;
+	  while (ISWHITE (line.ptr[l]) && l < line.len)
+	    l++;
+
+	  if (label_in.len)
+	    {
+	      int do_assigns;
+
+	      /* Munge the label, unless this is EQU or ASSIGN.  */
+	      do_assigns = 1;
+	      if (l < line.len
+		  && (line.ptr[l] == '.' || alternate || mri))
+		{
+		  int lx = l;
+
+		  if (line.ptr[lx] == '.')
+		    ++lx;
+		  if (lx + 3 <= line.len
+		      && strncasecmp ("EQU", line.ptr + lx, 3) == 0
+		      && (lx + 3 == line.len
+			  || ! ISFIRSTCHAR (line.ptr[lx + 3])))
+		    do_assigns = 0;
+		  else if (lx + 6 <= line.len
+			   && strncasecmp ("ASSIGN", line.ptr + lx, 6) == 0
+			   && (lx + 6 == line.len
+			       || ! ISFIRSTCHAR (line.ptr[lx + 6])))
+		    do_assigns = 0;
+		}
+
+	      if (do_assigns)
+		process_assigns (0, &label_in, &label);
+	      else
+		sb_add_sb (&label, &label_in);
+	    }
+
+	  if (l < line.len)
+	    {
+	      if (process_pseudo_op (l, &line, &acc))
+		{
+
+		}
+	      else if (condass_on ())
+		{
+		  if (macro_op (l, &line))
+		    {
+
+		    }
+		  else
+		    {
+		      {
+			if (label.len)
+			  {
+			    fprintf (outfile, "%s:\t", sb_name (&label));
+			  }
+			else
+			  fprintf (outfile, "\t");
+			sb_reset (&t1);
+			process_assigns (l, &line, &t1);
+			sb_reset (&t2);
+			change_base (0, &t1, &t2);
+			fprintf (outfile, "%s\n", sb_name (&t2));
+		      }
+		    }
+		}
+	    }
+	  else
+	    {
+	      /* Only a label on this line.  */
+	      if (label.len && condass_on ())
+		{
+		  fprintf (outfile, "%s:\n", sb_name (&label));
+		}
+	    }
+	}
+
+      if (had_end)
+	break;
+      sb_reset (&line);
+      more = get_line (&line);
+    }
+
+  if (!had_end && !mri)
+    WARNING ((stderr, _("END missing from end of file.\n")));
+}
+
+static void
+free_old_entry (ptr)
+     hash_entry *ptr;
+{
+  if (ptr)
+    {
+      if (ptr->type == hash_string)
+	sb_kill (&ptr->value.s);
+    }
+}
+
+/* name: .ASSIGNA <value>  */
+
+static void
+do_assigna (idx, in)
+     int idx;
+     sb *in;
+{
+  sb tmp;
+  int val;
+  sb_new (&tmp);
+
+  process_assigns (idx, in, &tmp);
+  idx = exp_get_abs (_(".ASSIGNA needs constant expression argument.\n"), 0, &tmp, &val);
+
+  if (!label.len)
+    {
+      ERROR ((stderr, _(".ASSIGNA without label.\n")));
+    }
+  else
+    {
+      hash_entry *ptr = hash_create (&vars, &label);
+      free_old_entry (ptr);
+      ptr->type = hash_integer;
+      ptr->value.i = val;
+    }
+  sb_kill (&tmp);
+}
+
+/* name: .ASSIGNC <string>  */
+
+static void
+do_assignc (idx, in)
+     int idx;
+     sb *in;
+{
+  sb acc;
+  sb_new (&acc);
+  idx = getstring (idx, in, &acc);
+
+  if (!label.len)
+    {
+      ERROR ((stderr, _(".ASSIGNS without label.\n")));
+    }
+  else
+    {
+      hash_entry *ptr = hash_create (&vars, &label);
+      free_old_entry (ptr);
+      ptr->type = hash_string;
+      sb_new (&ptr->value.s);
+      sb_add_sb (&ptr->value.s, &acc);
+    }
+  sb_kill (&acc);
+}
+
+/* name: .REG (reg)  */
+
+static void
+do_reg (idx, in)
+     int idx;
+     sb *in;
+{
+  /* Remove reg stuff from inside parens.  */
+  sb what;
+  if (!mri)
+    idx = skip_openp (idx, in);
+  else
+    idx = sb_skip_white (idx, in);
+  sb_new (&what);
+  while (idx < in->len
+	 && (mri
+	     ? ! eol (idx, in)
+	     : in->ptr[idx] != ')'))
+    {
+      sb_add_char (&what, in->ptr[idx]);
+      idx++;
+    }
+  hash_add_to_string_table (&assign_hash_table, &label, &what, 1);
+  sb_kill (&what);
+}
+
+static int
+condass_lookup_name (inbuf, idx, out, warn)
+     sb *inbuf;
+     int idx;
+     sb *out;
+     int warn;
+{
+  hash_entry *ptr;
+  sb condass_acc;
+  sb_new (&condass_acc);
+
+  while (idx < inbuf->len
+	 && ISNEXTCHAR (inbuf->ptr[idx]))
+    {
+      sb_add_char (&condass_acc, inbuf->ptr[idx++]);
+    }
+
+  if (inbuf->ptr[idx] == '\'')
+    idx++;
+  ptr = hash_lookup (&vars, &condass_acc);
+
+  if (!ptr)
+    {
+      if (warn)
+	{
+	  WARNING ((stderr, _("Can't find preprocessor variable %s.\n"), sb_name (&condass_acc)));
+	}
+      else
+	{
+	  sb_add_string (out, "0");
+	}
+    }
+  else
+    {
+      if (ptr->type == hash_integer)
+	{
+	  char buffer[30];
+	  sprintf (buffer, "%d", ptr->value.i);
+	  sb_add_string (out, buffer);
+	}
+      else
+	{
+	  sb_add_sb (out, &ptr->value.s);
+	}
+    }
+  sb_kill (&condass_acc);
+  return idx;
+}
+
+#define EQ 1
+#define NE 2
+#define GE 3
+#define LT 4
+#define LE 5
+#define GT 6
+#define NEVER 7
+
+static int
+whatcond (idx, in, val)
+     int idx;
+     sb *in;
+     int *val;
+{
+  int cond;
+
+  idx = sb_skip_white (idx, in);
+  cond = NEVER;
+  if (idx + 1 < in->len)
+    {
+      char *p;
+      char a, b;
+
+      p = in->ptr + idx;
+      a = TOUPPER (p[0]);
+      b = TOUPPER (p[1]);
+      if (a == 'E' && b == 'Q')
+	cond = EQ;
+      else if (a == 'N' && b == 'E')
+	cond = NE;
+      else if (a == 'L' && b == 'T')
+	cond = LT;
+      else if (a == 'L' && b == 'E')
+	cond = LE;
+      else if (a == 'G' && b == 'T')
+	cond = GT;
+      else if (a == 'G' && b == 'E')
+	cond = GE;
+    }
+  if (cond == NEVER)
+    {
+      ERROR ((stderr, _("Comparison operator must be one of EQ, NE, LT, LE, GT or GE.\n")));
+      cond = NEVER;
+    }
+  idx = sb_skip_white (idx + 2, in);
+  *val = cond;
+  return idx;
+}
+
+static int
+istrue (idx, in)
+     int idx;
+     sb *in;
+{
+  int res;
+  sb acc_a;
+  sb cond;
+  sb acc_b;
+  sb_new (&acc_a);
+  sb_new (&cond);
+  sb_new (&acc_b);
+  idx = sb_skip_white (idx, in);
+
+  if (in->ptr[idx] == '"')
+    {
+      int cond;
+      int same;
+      /* This is a string comparision.  */
+      idx = getstring (idx, in, &acc_a);
+      idx = whatcond (idx, in, &cond);
+      idx = getstring (idx, in, &acc_b);
+      same = acc_a.len == acc_b.len
+	&& (strncmp (acc_a.ptr, acc_b.ptr, acc_a.len) == 0);
+
+      if (cond != EQ && cond != NE)
+	{
+	  ERROR ((stderr, _("Comparison operator for strings must be EQ or NE\n")));
+	  res = 0;
+	}
+      else
+	res = (cond != EQ) ^ same;
+    }
+  else
+    /* This is a numeric expression.  */
+    {
+      int vala;
+      int valb;
+      int cond;
+      idx = exp_get_abs (_("Conditional operator must have absolute operands.\n"), idx, in, &vala);
+      idx = whatcond (idx, in, &cond);
+      idx = sb_skip_white (idx, in);
+      if (in->ptr[idx] == '"')
+	{
+	  WARNING ((stderr, _("String compared against expression.\n")));
+	  res = 0;
+	}
+      else
+	{
+	  idx = exp_get_abs (_("Conditional operator must have absolute operands.\n"), idx, in, &valb);
+	  switch (cond)
+	    {
+	    default:
+	      res = 42;
+	      break;
+	    case EQ:
+	      res = vala == valb;
+	      break;
+	    case NE:
+	      res = vala != valb;
+	      break;
+	    case LT:
+	      res = vala < valb;
+	      break;
+	    case LE:
+	      res = vala <= valb;
+	      break;
+	    case GT:
+	      res = vala > valb;
+	      break;
+	    case GE:
+	      res = vala >= valb;
+	      break;
+	    case NEVER:
+	      res = 0;
+	      break;
+	    }
+	}
+    }
+
+  sb_kill (&acc_a);
+  sb_kill (&cond);
+  sb_kill (&acc_b);
+  return res;
+}
+
+/* .AIF  */
+
+static void
+do_aif (idx, in)
+     int idx;
+     sb *in;
+{
+  if (ifi >= IFNESTING)
+    {
+      FATAL ((stderr, _("AIF nesting unreasonable.\n")));
+    }
+  ifi++;
+  ifstack[ifi].on = ifstack[ifi - 1].on ? istrue (idx, in) : 0;
+  ifstack[ifi].hadelse = 0;
+}
+
+/* .AELSE  */
+
+static void
+do_aelse ()
+{
+  ifstack[ifi].on = ifstack[ifi - 1].on ? !ifstack[ifi].on : 0;
+  if (ifstack[ifi].hadelse)
+    {
+      ERROR ((stderr, _("Multiple AELSEs in AIF.\n")));
+    }
+  ifstack[ifi].hadelse = 1;
+}
+
+/* .AENDI  */
+
+static void
+do_aendi ()
+{
+  if (ifi != 0)
+    {
+      ifi--;
+    }
+  else
+    {
+      ERROR ((stderr, _("AENDI without AIF.\n")));
+    }
+}
+
+static int
+condass_on ()
+{
+  return ifstack[ifi].on;
+}
+
+/* MRI IFEQ, IFNE, IFLT, IFLE, IFGE, IFGT.  */
+
+static void
+do_if (idx, in, cond)
+     int idx;
+     sb *in;
+     int cond;
+{
+  int val;
+  int res;
+
+  if (ifi >= IFNESTING)
+    {
+      FATAL ((stderr, _("IF nesting unreasonable.\n")));
+    }
+
+  idx = exp_get_abs (_("Conditional operator must have absolute operands.\n"),
+		     idx, in, &val);
+  switch (cond)
+    {
+    default:
+    case EQ: res = val == 0; break;
+    case NE: res = val != 0; break;
+    case LT: res = val <  0; break;
+    case LE: res = val <= 0; break;
+    case GE: res = val >= 0; break;
+    case GT: res = val >  0; break;
+    }
+
+  ifi++;
+  ifstack[ifi].on = ifstack[ifi - 1].on ? res : 0;
+  ifstack[ifi].hadelse = 0;
+}
+
+/* Get a string for the MRI IFC or IFNC pseudo-ops.  */
+
+static int
+get_mri_string (idx, in, val, terminator)
+     int idx;
+     sb *in;
+     sb *val;
+     int terminator;
+{
+  idx = sb_skip_white (idx, in);
+
+  if (idx < in->len
+      && in->ptr[idx] == '\'')
+    {
+      sb_add_char (val, '\'');
+      for (++idx; idx < in->len; ++idx)
+	{
+	  sb_add_char (val, in->ptr[idx]);
+	  if (in->ptr[idx] == '\'')
+	    {
+	      ++idx;
+	      if (idx >= in->len
+		  || in->ptr[idx] != '\'')
+		break;
+	    }
+	}
+      idx = sb_skip_white (idx, in);
+    }
+  else
+    {
+      int i;
+
+      while (idx < in->len
+	     && in->ptr[idx] != terminator)
+	{
+	  sb_add_char (val, in->ptr[idx]);
+	  ++idx;
+	}
+      i = val->len - 1;
+      while (i >= 0 && ISWHITE (val->ptr[i]))
+	--i;
+      val->len = i + 1;
+    }
+
+  return idx;
+}
+
+/* MRI IFC, IFNC  */
+
+static void
+do_ifc (idx, in, ifnc)
+     int idx;
+     sb *in;
+     int ifnc;
+{
+  sb first;
+  sb second;
+  int res;
+
+  if (ifi >= IFNESTING)
+    {
+      FATAL ((stderr, _("IF nesting unreasonable.\n")));
+    }
+
+  sb_new (&first);
+  sb_new (&second);
+
+  idx = get_mri_string (idx, in, &first, ',');
+
+  if (idx >= in->len || in->ptr[idx] != ',')
+    {
+      ERROR ((stderr, _("Bad format for IF or IFNC.\n")));
+      return;
+    }
+
+  idx = get_mri_string (idx + 1, in, &second, ';');
+
+  res = (first.len == second.len
+	 && strncmp (first.ptr, second.ptr, first.len) == 0);
+  res ^= ifnc;
+
+  ifi++;
+  ifstack[ifi].on = ifstack[ifi - 1].on ? res : 0;
+  ifstack[ifi].hadelse = 0;
+}
+
+/* .ENDR  */
+
+static void
+do_aendr ()
+{
+  if (!mri)
+    ERROR ((stderr, _("AENDR without a AREPEAT.\n")));
+  else
+    ERROR ((stderr, _("ENDR without a REPT.\n")));
+}
+
+/* .AWHILE  */
+
+static void
+do_awhile (idx, in)
+     int idx;
+     sb *in;
+{
+  int line = linecount ();
+  sb exp;
+  sb sub;
+  int doit;
+
+  sb_new (&sub);
+  sb_new (&exp);
+
+  process_assigns (idx, in, &exp);
+  doit = istrue (0, &exp);
+
+  if (! buffer_and_nest ("AWHILE", "AENDW", &sub, get_line))
+    FATAL ((stderr, _("AWHILE without a AENDW at %d.\n"), line - 1));
+
+  /* Turn
+     	.AWHILE exp
+	     foo
+	.AENDW
+     into
+        foo
+	.AWHILE exp
+	foo
+	.ENDW
+  */
+
+  if (doit)
+    {
+      int index = include_next_index ();
+
+      sb copy;
+      sb_new (&copy);
+      sb_add_sb (&copy, &sub);
+      sb_add_sb (&copy, in);
+      sb_add_string (&copy, "\n");
+      sb_add_sb (&copy, &sub);
+      sb_add_string (&copy, "\t.AENDW\n");
+      /* Push another WHILE.  */
+      include_buf (&exp, &copy, include_while, index);
+      sb_kill (&copy);
+    }
+  sb_kill (&exp);
+  sb_kill (&sub);
+}
+
+/* .AENDW  */
+
+static void
+do_aendw ()
+{
+  ERROR ((stderr, _("AENDW without a AENDW.\n")));
+}
+
+/* .EXITM
+
+   Pop things off the include stack until the type and index changes.  */
+
+static void
+do_exitm ()
+{
+  include_type type = sp->type;
+  if (type == include_repeat
+      || type == include_while
+      || type == include_macro)
+    {
+      int index = sp->index;
+      include_pop ();
+      while (sp->index == index
+	     && sp->type == type)
+	{
+	  include_pop ();
+	}
+    }
+}
+
+/* .AREPEAT  */
+
+static void
+do_arepeat (idx, in)
+     int idx;
+     sb *in;
+{
+  int line = linecount ();
+  sb exp;			/* Buffer with expression in it.  */
+  sb copy;			/* Expanded repeat block.  */
+  sb sub;			/* Contents of AREPEAT.  */
+  int rc;
+  int ret;
+  char buffer[30];
+
+  sb_new (&exp);
+  sb_new (&copy);
+  sb_new (&sub);
+  process_assigns (idx, in, &exp);
+  idx = exp_get_abs (_("AREPEAT must have absolute operand.\n"), 0, &exp, &rc);
+  if (!mri)
+    ret = buffer_and_nest ("AREPEAT", "AENDR", &sub, get_line);
+  else
+    ret = buffer_and_nest ("REPT", "ENDR", &sub, get_line);
+  if (! ret)
+    FATAL ((stderr, _("AREPEAT without a AENDR at %d.\n"), line - 1));
+  if (rc > 0)
+    {
+      /* Push back the text following the repeat, and another repeat block
+	 so
+	 .AREPEAT 20
+	 foo
+	 .AENDR
+	 gets turned into
+	 foo
+	 .AREPEAT 19
+	 foo
+	 .AENDR
+      */
+      int index = include_next_index ();
+      sb_add_sb (&copy, &sub);
+      if (rc > 1)
+	{
+	  if (!mri)
+	    sprintf (buffer, "\t.AREPEAT	%d\n", rc - 1);
+	  else
+	    sprintf (buffer, "\tREPT	%d\n", rc - 1);
+	  sb_add_string (&copy, buffer);
+	  sb_add_sb (&copy, &sub);
+	  if (!mri)
+	    sb_add_string (&copy, "	.AENDR\n");
+	  else
+	    sb_add_string (&copy, "	ENDR\n");
+	}
+
+      include_buf (&exp, &copy, include_repeat, index);
+    }
+  sb_kill (&exp);
+  sb_kill (&sub);
+  sb_kill (&copy);
+}
+
+/* .ENDM  */
+
+static void
+do_endm ()
+{
+  ERROR ((stderr, _(".ENDM without a matching .MACRO.\n")));
+}
+
+/* MRI IRP pseudo-op.  */
+
+static void
+do_irp (idx, in, irpc)
+     int idx;
+     sb *in;
+     int irpc;
+{
+  const char *err;
+  sb out;
+
+  sb_new (&out);
+
+  err = expand_irp (irpc, idx, in, &out, get_line, comment_char);
+  if (err != NULL)
+    ERROR ((stderr, "%s\n", err));
+
+  fprintf (outfile, "%s", sb_terminate (&out));
+
+  sb_kill (&out);
+}
+
+/* Macro processing.  */
+
+/* Parse off LOCAL n1, n2,... Invent a label name for it.  */
+
+static void
+do_local (idx, line)
+     int idx ATTRIBUTE_UNUSED;
+     sb *line ATTRIBUTE_UNUSED;
+{
+  ERROR ((stderr, _("LOCAL outside of MACRO")));
+}
+
+static void
+do_macro (idx, in)
+     int idx;
+     sb *in;
+{
+  const char *err;
+  int line = linecount ();
+
+  err = define_macro (idx, in, &label, get_line, "[gasp]", line, (const char **) NULL);
+  if (err != NULL)
+    ERROR ((stderr, _("macro at line %d: %s\n"), line - 1, err));
+}
+
+static int
+macro_op (idx, in)
+     int idx;
+     sb *in;
+{
+  const char *err;
+  sb out;
+  sb name;
+
+  if (! macro_defined)
+    return 0;
+
+  sb_terminate (in);
+  if (! check_macro (in->ptr + idx, &out, comment_char, &err, NULL))
+    return 0;
+
+  if (err != NULL)
+    ERROR ((stderr, "%s\n", err));
+
+  sb_new (&name);
+  sb_add_string (&name, _("macro expansion"));
+
+  include_buf (&name, &out, include_macro, include_next_index ());
+
+  sb_kill (&name);
+  sb_kill (&out);
+
+  return 1;
+}
+
+/* String handling.  */
+
+static int
+getstring (idx, in, acc)
+     int idx;
+     sb *in;
+     sb *acc;
+{
+  idx = sb_skip_white (idx, in);
+
+  while (idx < in->len
+	 && (in->ptr[idx] == '"'
+	     || in->ptr[idx] == '<'
+	     || (in->ptr[idx] == '\'' && alternate)))
+    {
+      if (in->ptr[idx] == '<')
+	{
+	  if (alternate || mri)
+	    {
+	      int nest = 0;
+	      idx++;
+	      while ((in->ptr[idx] != '>' || nest)
+		     && idx < in->len)
+		{
+		  if (in->ptr[idx] == '!')
+		    {
+		      idx++;
+		      sb_add_char (acc, in->ptr[idx++]);
+		    }
+		  else
+		    {
+		      if (in->ptr[idx] == '>')
+			nest--;
+		      if (in->ptr[idx] == '<')
+			nest++;
+		      sb_add_char (acc, in->ptr[idx++]);
+		    }
+		}
+	      idx++;
+	    }
+	  else
+	    {
+	      int code;
+	      idx++;
+	      idx = exp_get_abs (_("Character code in string must be absolute expression.\n"),
+				 idx, in, &code);
+	      sb_add_char (acc, code);
+
+	      if (in->ptr[idx] != '>')
+		ERROR ((stderr, _("Missing > for character code.\n")));
+	      idx++;
+	    }
+	}
+      else if (in->ptr[idx] == '"' || in->ptr[idx] == '\'')
+	{
+	  char tchar = in->ptr[idx];
+	  idx++;
+	  while (idx < in->len)
+	    {
+	      if (alternate && in->ptr[idx] == '!')
+		{
+		  idx++;
+		  sb_add_char (acc, in->ptr[idx++]);
+		}
+	      else
+		{
+		  if (in->ptr[idx] == tchar)
+		    {
+		      idx++;
+		      if (idx >= in->len || in->ptr[idx] != tchar)
+			break;
+		    }
+		  sb_add_char (acc, in->ptr[idx]);
+		  idx++;
+		}
+	    }
+	}
+    }
+
+  return idx;
+}
+
+/* .SDATA[C|Z] <string>  */
+
+static void
+do_sdata (idx, in, type)
+     int idx;
+     sb *in;
+     int type;
+{
+  int nc = 0;
+  int pidx = -1;
+  sb acc;
+  sb_new (&acc);
+  fprintf (outfile, ".byte\t");
+
+  while (!eol (idx, in))
+    {
+      int i;
+      sb_reset (&acc);
+      idx = sb_skip_white (idx, in);
+      while (!eol (idx, in))
+	{
+	  pidx = idx = get_any_string (idx, in, &acc, 0, 1);
+	  if (type == 'c')
+	    {
+	      if (acc.len > 255)
+		{
+		  ERROR ((stderr, _("string for SDATAC longer than 255 characters (%d).\n"), acc.len));
+		}
+	      fprintf (outfile, "%d", acc.len);
+	      nc = 1;
+	    }
+
+	  for (i = 0; i < acc.len; i++)
+	    {
+	      if (nc)
+		{
+		  fprintf (outfile, ",");
+		}
+	      fprintf (outfile, "%d", acc.ptr[i]);
+	      nc = 1;
+	    }
+
+	  if (type == 'z')
+	    {
+	      if (nc)
+		fprintf (outfile, ",");
+	      fprintf (outfile, "0");
+	    }
+	  idx = sb_skip_comma (idx, in);
+	  if (idx == pidx)
+	    break;
+	}
+      if (!alternate && in->ptr[idx] != ',' && idx != in->len)
+	{
+	  fprintf (outfile, "\n");
+	  ERROR ((stderr, _("illegal character in SDATA line (0x%x).\n"),
+		  in->ptr[idx]));
+	  break;
+	}
+      idx++;
+    }
+  sb_kill (&acc);
+  fprintf (outfile, "\n");
+}
+
+/* .SDATAB <count> <string>  */
+
+static void
+do_sdatab (idx, in)
+     int idx;
+     sb *in;
+{
+  int repeat;
+  int i;
+  sb acc;
+  sb_new (&acc);
+
+  idx = exp_get_abs (_("Must have absolute SDATAB repeat count.\n"), idx, in, &repeat);
+  if (repeat <= 0)
+    {
+      ERROR ((stderr, _("Must have positive SDATAB repeat count (%d).\n"), repeat));
+      repeat = 1;
+    }
+
+  idx = sb_skip_comma (idx, in);
+  idx = getstring (idx, in, &acc);
+
+  for (i = 0; i < repeat; i++)
+    {
+      if (i)
+	fprintf (outfile, "\t");
+      fprintf (outfile, ".byte\t");
+      sb_print (outfile, &acc);
+      fprintf (outfile, "\n");
+    }
+  sb_kill (&acc);
+
+}
+
+static int
+new_file (name)
+     const char *name;
+{
+  FILE *newone = fopen (name, "r");
+  if (!newone)
+    return 0;
+
+  if (isp == MAX_INCLUDES)
+    FATAL ((stderr, _("Unreasonable include depth (%ld).\n"), (long) isp));
+
+  sp++;
+  sp->handle = newone;
+
+  sb_new (&sp->name);
+  sb_add_string (&sp->name, name);
+
+  sp->linecount = 1;
+  sp->pushback_index = 0;
+  sp->type = include_file;
+  sp->index = 0;
+  sb_new (&sp->pushback);
+  return 1;
+}
+
+static void
+do_include (idx, in)
+     int idx;
+     sb *in;
+{
+  sb t;
+  sb cat;
+  include_path *includes;
+
+  sb_new (&t);
+  sb_new (&cat);
+
+  if (! mri)
+    idx = getstring (idx, in, &t);
+  else
+    {
+      idx = sb_skip_white (idx, in);
+      while (idx < in->len && ! ISWHITE (in->ptr[idx]))
+	{
+	  sb_add_char (&t, in->ptr[idx]);
+	  ++idx;
+	}
+    }
+
+  for (includes = paths_head; includes; includes = includes->next)
+    {
+      sb_reset (&cat);
+      sb_add_sb (&cat, &includes->path);
+      sb_add_char (&cat, '/');
+      sb_add_sb (&cat, &t);
+      if (new_file (sb_name (&cat)))
+	{
+	  break;
+	}
+    }
+  if (!includes)
+    {
+      if (! new_file (sb_name (&t)))
+	FATAL ((stderr, _("Can't open include file `%s'.\n"), sb_name (&t)));
+    }
+  sb_kill (&cat);
+  sb_kill (&t);
+}
+
+static void
+include_pop ()
+{
+  if (sp != include_stack)
+    {
+      if (sp->handle)
+	fclose (sp->handle);
+      sp--;
+    }
+}
+
+/* Get the next character from the include stack.  If there's anything
+   in the pushback buffer, take that first.  If we're at eof, pop from
+   the stack and try again.  Keep the linecount up to date.  */
+
+static int
+get ()
+{
+  int r;
+
+  if (sp->pushback.len != sp->pushback_index)
+    {
+      r = (char) (sp->pushback.ptr[sp->pushback_index++]);
+      /* When they've all gone, reset the pointer.  */
+      if (sp->pushback_index == sp->pushback.len)
+	{
+	  sp->pushback.len = 0;
+	  sp->pushback_index = 0;
+	}
+    }
+  else if (sp->handle)
+    {
+      r = getc (sp->handle);
+    }
+  else
+    r = EOF;
+
+  if (r == EOF && isp)
+    {
+      include_pop ();
+      r = get ();
+      while (r == EOF && isp)
+	{
+	  include_pop ();
+	  r = get ();
+	}
+      return r;
+    }
+  if (r == '\n')
+    {
+      sp->linecount++;
+    }
+
+  return r;
+}
+
+static int
+linecount ()
+{
+  return sp->linecount;
+}
+
+static int
+include_next_index ()
+{
+  static int index;
+  if (!unreasonable
+      && index > MAX_REASONABLE)
+    FATAL ((stderr, _("Unreasonable expansion (-u turns off check).\n")));
+  return ++index;
+}
+
+/* Initialize the chartype vector.  */
+
+static void
+chartype_init ()
+{
+  int x;
+  for (x = 0; x < 256; x++)
+    {
+      if (ISALPHA (x) || x == '_' || x == '$')
+	chartype[x] |= FIRSTBIT;
+
+      if (mri && x == '.')
+	chartype[x] |= FIRSTBIT;
+
+      if (ISDIGIT (x) || ISALPHA (x) || x == '_' || x == '$')
+	chartype[x] |= NEXTBIT;
+
+      if (x == ' ' || x == '\t' || x == ',' || x == '"' || x == ';'
+	  || x == '"' || x == '<' || x == '>' || x == ')' || x == '(')
+	chartype[x] |= SEPBIT;
+
+      if (x == 'b' || x == 'B'
+	  || x == 'q' || x == 'Q'
+	  || x == 'h' || x == 'H'
+	  || x == 'd' || x == 'D')
+	chartype [x] |= BASEBIT;
+
+      if (x == ' ' || x == '\t')
+	chartype[x] |= WHITEBIT;
+
+      if (x == comment_char)
+	chartype[x] |= COMMENTBIT;
+    }
+}
+
+/* What to do with all the keywords.  */
+#define PROCESS 	0x1000  /* Run substitution over the line.  */
+#define LAB		0x2000  /* Spit out the label.  */
+
+#define K_EQU 		(PROCESS|1)
+#define K_ASSIGN 	(PROCESS|2)
+#define K_REG 		(PROCESS|3)
+#define K_ORG 		(PROCESS|4)
+#define K_RADIX 	(PROCESS|5)
+#define K_DATA 		(LAB|PROCESS|6)
+#define K_DATAB 	(LAB|PROCESS|7)
+#define K_SDATA 	(LAB|PROCESS|8)
+#define K_SDATAB 	(LAB|PROCESS|9)
+#define K_SDATAC 	(LAB|PROCESS|10)
+#define K_SDATAZ	(LAB|PROCESS|11)
+#define K_RES 		(LAB|PROCESS|12)
+#define K_SRES 		(LAB|PROCESS|13)
+#define K_SRESC 	(LAB|PROCESS|14)
+#define K_SRESZ 	(LAB|PROCESS|15)
+#define K_EXPORT 	(LAB|PROCESS|16)
+#define K_GLOBAL 	(LAB|PROCESS|17)
+#define K_PRINT 	(LAB|PROCESS|19)
+#define K_FORM 		(LAB|PROCESS|20)
+#define K_HEADING	(LAB|PROCESS|21)
+#define K_PAGE		(LAB|PROCESS|22)
+#define K_IMPORT	(LAB|PROCESS|23)
+#define K_PROGRAM	(LAB|PROCESS|24)
+#define K_END		(PROCESS|25)
+#define K_INCLUDE	(PROCESS|26)
+#define K_IGNORED	(PROCESS|27)
+#define K_ASSIGNA	(PROCESS|28)
+#define K_ASSIGNC	(29)
+#define K_AIF		(PROCESS|30)
+#define K_AELSE		(PROCESS|31)
+#define K_AENDI		(PROCESS|32)
+#define K_AREPEAT	(PROCESS|33)
+#define K_AENDR		(PROCESS|34)
+#define K_AWHILE	(35)
+#define K_AENDW		(PROCESS|36)
+#define K_EXITM		(37)
+#define K_MACRO		(PROCESS|38)
+#define K_ENDM		(39)
+#define K_ALIGN		(PROCESS|LAB|40)
+#define K_ALTERNATE     (41)
+#define K_DB		(LAB|PROCESS|42)
+#define K_DW		(LAB|PROCESS|43)
+#define K_DL		(LAB|PROCESS|44)
+#define K_LOCAL		(45)
+#define K_IFEQ		(PROCESS|46)
+#define K_IFNE		(PROCESS|47)
+#define K_IFLT		(PROCESS|48)
+#define K_IFLE		(PROCESS|49)
+#define K_IFGE		(PROCESS|50)
+#define K_IFGT		(PROCESS|51)
+#define K_IFC		(PROCESS|52)
+#define K_IFNC		(PROCESS|53)
+#define K_IRP		(PROCESS|54)
+#define K_IRPC		(PROCESS|55)
+
+struct keyword {
+  char *name;
+  int code;
+  int extra;
+};
+
+static struct keyword kinfo[] = {
+  { "EQU", K_EQU, 0 },
+  { "ALTERNATE", K_ALTERNATE, 0 },
+  { "ASSIGN", K_ASSIGN, 0 },
+  { "REG", K_REG, 0 },
+  { "ORG", K_ORG, 0 },
+  { "RADIX", K_RADIX, 0 },
+  { "DATA", K_DATA, 0 },
+  { "DB", K_DB, 0 },
+  { "DW", K_DW, 0 },
+  { "DL", K_DL, 0 },
+  { "DATAB", K_DATAB, 0 },
+  { "SDATA", K_SDATA, 0 },
+  { "SDATAB", K_SDATAB, 0 },
+  { "SDATAZ", K_SDATAZ, 0 },
+  { "SDATAC", K_SDATAC, 0 },
+  { "RES", K_RES, 0 },
+  { "SRES", K_SRES, 0 },
+  { "SRESC", K_SRESC, 0 },
+  { "SRESZ", K_SRESZ, 0 },
+  { "EXPORT", K_EXPORT, 0 },
+  { "GLOBAL", K_GLOBAL, 0 },
+  { "PRINT", K_PRINT, 0 },
+  { "FORM", K_FORM, 0 },
+  { "HEADING", K_HEADING, 0 },
+  { "PAGE", K_PAGE, 0 },
+  { "PROGRAM", K_IGNORED, 0 },
+  { "END", K_END, 0 },
+  { "INCLUDE", K_INCLUDE, 0 },
+  { "ASSIGNA", K_ASSIGNA, 0 },
+  { "ASSIGNC", K_ASSIGNC, 0 },
+  { "AIF", K_AIF, 0 },
+  { "AELSE", K_AELSE, 0 },
+  { "AENDI", K_AENDI, 0 },
+  { "AREPEAT", K_AREPEAT, 0 },
+  { "AENDR", K_AENDR, 0 },
+  { "EXITM", K_EXITM, 0 },
+  { "MACRO", K_MACRO, 0 },
+  { "ENDM", K_ENDM, 0 },
+  { "AWHILE", K_AWHILE, 0 },
+  { "ALIGN", K_ALIGN, 0 },
+  { "AENDW", K_AENDW, 0 },
+  { "ALTERNATE", K_ALTERNATE, 0 },
+  { "LOCAL", K_LOCAL, 0 },
+  { NULL, 0, 0 }
+};
+
+/* Although the conditional operators are handled by gas, we need to
+   handle them here as well, in case they are used in a recursive
+   macro to end the recursion.  */
+
+static struct keyword mrikinfo[] = {
+  { "IFEQ", K_IFEQ, 0 },
+  { "IFNE", K_IFNE, 0 },
+  { "IFLT", K_IFLT, 0 },
+  { "IFLE", K_IFLE, 0 },
+  { "IFGE", K_IFGE, 0 },
+  { "IFGT", K_IFGT, 0 },
+  { "IFC", K_IFC, 0 },
+  { "IFNC", K_IFNC, 0 },
+  { "ELSEC", K_AELSE, 0 },
+  { "ENDC", K_AENDI, 0 },
+  { "MEXIT", K_EXITM, 0 },
+  { "REPT", K_AREPEAT, 0 },
+  { "IRP", K_IRP, 0 },
+  { "IRPC", K_IRPC, 0 },
+  { "ENDR", K_AENDR, 0 },
+  { NULL, 0, 0 }
+};
+
+/* Look for a pseudo op on the line. If one's there then call
+   its handler.  */
+
+static int
+process_pseudo_op (idx, line, acc)
+     int idx;
+     sb *line;
+     sb *acc;
+{
+  int oidx = idx;
+
+  if (line->ptr[idx] == '.' || alternate || mri)
+    {
+      /* Scan forward and find pseudo name.  */
+      char *in;
+      hash_entry *ptr;
+
+      char *s;
+      char *e;
+      if (line->ptr[idx] == '.')
+	idx++;
+      in = line->ptr + idx;
+      s = in;
+      e = s;
+      sb_reset (acc);
+
+      while (idx < line->len && *e && ISFIRSTCHAR (*e))
+	{
+	  sb_add_char (acc, *e);
+	  e++;
+	  idx++;
+	}
+
+      ptr = hash_lookup (&keyword_hash_table, acc);
+
+      if (!ptr)
+	{
+#if 0
+	  /* This one causes lots of pain when trying to preprocess
+	     ordinary code.  */
+	  WARNING ((stderr, _("Unrecognised pseudo op `%s'.\n"),
+		    sb_name (acc)));
+#endif
+	  return 0;
+	}
+      if (ptr->value.i & LAB)
+	{
+	  /* Output the label.  */
+	  if (label.len)
+	    {
+	      fprintf (outfile, "%s:\t", sb_name (&label));
+	    }
+	  else
+	    fprintf (outfile, "\t");
+	}
+
+      if (mri && ptr->value.i == K_END)
+	{
+	  sb t;
+
+	  sb_new (&t);
+	  sb_add_buffer (&t, line->ptr + oidx, idx - oidx);
+	  fprintf (outfile, "\t%s", sb_name (&t));
+	  sb_kill (&t);
+	}
+
+      if (ptr->value.i & PROCESS)
+	{
+	  /* Polish the rest of the line before handling the pseudo op.  */
+#if 0
+	  strip_comments (line);
+#endif
+	  sb_reset (acc);
+	  process_assigns (idx, line, acc);
+	  sb_reset (line);
+	  change_base (0, acc, line);
+	  idx = 0;
+	}
+      if (!condass_on ())
+	{
+	  switch (ptr->value.i)
+	    {
+	    case K_AIF:
+	      do_aif (idx, line);
+	      break;
+	    case K_AELSE:
+	      do_aelse ();
+	      break;
+	    case K_AENDI:
+	      do_aendi ();
+	      break;
+	    }
+	  return 1;
+	}
+      else
+	{
+	  switch (ptr->value.i)
+	    {
+	    case K_ALTERNATE:
+	      alternate = 1;
+	      macro_init (1, mri, 0, exp_get_abs);
+	      return 1;
+	    case K_AELSE:
+	      do_aelse ();
+	      return 1;
+	    case K_AENDI:
+	      do_aendi ();
+	      return 1;
+	    case K_ORG:
+	      ERROR ((stderr, _("ORG command not allowed.\n")));
+	      break;
+	    case K_RADIX:
+	      do_radix (line);
+	      return 1;
+	    case K_DB:
+	      do_data (idx, line, 1);
+	      return 1;
+	    case K_DW:
+	      do_data (idx, line, 2);
+	      return 1;
+	    case K_DL:
+	      do_data (idx, line, 4);
+	      return 1;
+	    case K_DATA:
+	      do_data (idx, line, 0);
+	      return 1;
+	    case K_DATAB:
+	      do_datab (idx, line);
+	      return 1;
+	    case K_SDATA:
+	      do_sdata (idx, line, 0);
+	      return 1;
+	    case K_SDATAB:
+	      do_sdatab (idx, line);
+	      return 1;
+	    case K_SDATAC:
+	      do_sdata (idx, line, 'c');
+	      return 1;
+	    case K_SDATAZ:
+	      do_sdata (idx, line, 'z');
+	      return 1;
+	    case K_ASSIGN:
+	      do_assign (0, 0, line);
+	      return 1;
+	    case K_AIF:
+	      do_aif (idx, line);
+	      return 1;
+	    case K_AREPEAT:
+	      do_arepeat (idx, line);
+	      return 1;
+	    case K_AENDW:
+	      do_aendw ();
+	      return 1;
+	    case K_AWHILE:
+	      do_awhile (idx, line);
+	      return 1;
+	    case K_AENDR:
+	      do_aendr ();
+	      return 1;
+	    case K_EQU:
+	      do_assign (1, idx, line);
+	      return 1;
+	    case K_ALIGN:
+	      do_align (idx, line);
+	      return 1;
+	    case K_RES:
+	      do_res (idx, line, 0);
+	      return 1;
+	    case K_SRES:
+	      do_res (idx, line, 's');
+	      return 1;
+	    case K_INCLUDE:
+	      do_include (idx, line);
+	      return 1;
+	    case K_LOCAL:
+	      do_local (idx, line);
+	      return 1;
+	    case K_MACRO:
+	      do_macro (idx, line);
+	      return 1;
+	    case K_ENDM:
+	      do_endm ();
+	      return 1;
+	    case K_SRESC:
+	      do_res (idx, line, 'c');
+	      return 1;
+	    case K_PRINT:
+	      do_print (idx, line);
+	      return 1;
+	    case K_FORM:
+	      do_form (idx, line);
+	      return 1;
+	    case K_HEADING:
+	      do_heading (idx, line);
+	      return 1;
+	    case K_PAGE:
+	      do_page ();
+	      return 1;
+	    case K_GLOBAL:
+	    case K_EXPORT:
+	      do_export (line);
+	      return 1;
+	    case K_IMPORT:
+	      return 1;
+	    case K_SRESZ:
+	      do_res (idx, line, 'z');
+	      return 1;
+	    case K_IGNORED:
+	      return 1;
+	    case K_END:
+	      do_end (line);
+	      return 1;
+	    case K_ASSIGNA:
+	      do_assigna (idx, line);
+	      return 1;
+	    case K_ASSIGNC:
+	      do_assignc (idx, line);
+	      return 1;
+	    case K_EXITM:
+	      do_exitm ();
+	      return 1;
+	    case K_REG:
+	      do_reg (idx, line);
+	      return 1;
+	    case K_IFEQ:
+	      do_if (idx, line, EQ);
+	      return 1;
+	    case K_IFNE:
+	      do_if (idx, line, NE);
+	      return 1;
+	    case K_IFLT:
+	      do_if (idx, line, LT);
+	      return 1;
+	    case K_IFLE:
+	      do_if (idx, line, LE);
+	      return 1;
+	    case K_IFGE:
+	      do_if (idx, line, GE);
+	      return 1;
+	    case K_IFGT:
+	      do_if (idx, line, GT);
+	      return 1;
+	    case K_IFC:
+	      do_ifc (idx, line, 0);
+	      return 1;
+	    case K_IFNC:
+	      do_ifc (idx, line, 1);
+	      return 1;
+	    case K_IRP:
+	      do_irp (idx, line, 0);
+	      return 1;
+	    case K_IRPC:
+	      do_irp (idx, line, 1);
+	      return 1;
+	    }
+	}
+    }
+  return 0;
+}
+
+/* Add a keyword to the hash table.  */
+
+static void
+add_keyword (name, code)
+     const char *name;
+     int code;
+{
+  sb label;
+  int j;
+
+  sb_new (&label);
+  sb_add_string (&label, name);
+
+  hash_add_to_int_table (&keyword_hash_table, &label, code);
+
+  sb_reset (&label);
+  for (j = 0; name[j]; j++)
+    sb_add_char (&label, name[j] - 'A' + 'a');
+  hash_add_to_int_table (&keyword_hash_table, &label, code);
+
+  sb_kill (&label);
+}
+
+/* Build the keyword hash table - put each keyword in the table twice,
+   once upper and once lower case.  */
+
+static void
+process_init ()
+{
+  int i;
+
+  for (i = 0; kinfo[i].name; i++)
+    add_keyword (kinfo[i].name, kinfo[i].code);
+
+  if (mri)
+    {
+      for (i = 0; mrikinfo[i].name; i++)
+	add_keyword (mrikinfo[i].name, mrikinfo[i].code);
+    }
+}
+
+static void
+do_define (string)
+     const char *string;
+{
+  sb label;
+  int res = 1;
+  hash_entry *ptr;
+  sb_new (&label);
+
+  while (*string)
+    {
+      if (*string == '=')
+	{
+	  sb value;
+	  sb_new (&value);
+	  string++;
+	  while (*string)
+	    {
+	      sb_add_char (&value, *string);
+	      string++;
+	    }
+	  exp_get_abs (_("Invalid expression on command line.\n"),
+		       0, &value, &res);
+	  sb_kill (&value);
+	  break;
+	}
+      sb_add_char (&label, *string);
+
+      string++;
+    }
+
+  ptr = hash_create (&vars, &label);
+  free_old_entry (ptr);
+  ptr->type = hash_integer;
+  ptr->value.i = res;
+  sb_kill (&label);
+}
+
+char *program_name;
+
+/* The list of long options.  */
+static struct option long_options[] =
+{
+  { "alternate", no_argument, 0, 'a' },
+  { "include", required_argument, 0, 'I' },
+  { "commentchar", required_argument, 0, 'c' },
+  { "copysource", no_argument, 0, 's' },
+  { "debug", no_argument, 0, 'd' },
+  { "help", no_argument, 0, 'h' },
+  { "mri", no_argument, 0, 'M' },
+  { "output", required_argument, 0, 'o' },
+  { "print", no_argument, 0, 'p' },
+  { "unreasonable", no_argument, 0, 'u' },
+  { "version", no_argument, 0, 'v' },
+  { "define", required_argument, 0, 'd' },
+  { NULL, no_argument, 0, 0 }
+};
+
+/* Show a usage message and exit.  */
+static void
+show_usage (file, status)
+     FILE *file;
+     int status;
+{
+  fprintf (file, _("\
+Usage: %s \n\
+  [-a]      [--alternate]         enter alternate macro mode\n\
+  [-c char] [--commentchar char]  change the comment character from !\n\
+  [-d]      [--debug]             print some debugging info\n\
+  [-h]      [--help]              print this message\n\
+  [-M]      [--mri]               enter MRI compatibility mode\n\
+  [-o out]  [--output out]        set the output file\n\
+  [-p]      [--print]             print line numbers\n"), program_name);
+  fprintf (file, _("\
+  [-s]      [--copysource]        copy source through as comments \n\
+  [-u]      [--unreasonable]      allow unreasonable nesting\n\
+  [-v]      [--version]           print the program version\n\
+  [-Dname=value]                  create preprocessor variable called name, with value\n\
+  [-Ipath]                        add to include path list\n\
+  [in-file]\n"));
+  if (status == 0)
+    printf (_("Report bugs to %s\n"), REPORT_BUGS_TO);
+  exit (status);
+}
+
+/* Display a help message and exit.  */
+
+static void
+show_help ()
+{
+  printf (_("%s: Gnu Assembler Macro Preprocessor\n"), program_name);
+  show_usage (stdout, 0);
+}
+
+int main PARAMS ((int, char **));
+
+int
+main (argc, argv)
+     int argc;
+     char **argv;
+{
+  int opt;
+  char *out_name = 0;
+  sp = include_stack;
+
+  ifstack[0].on = 1;
+  ifi = 0;
+
+#if defined (HAVE_SETLOCALE) && defined (HAVE_LC_MESSAGES)
+  setlocale (LC_MESSAGES, "");
+#endif
+#if defined (HAVE_SETLOCALE)
+  setlocale (LC_CTYPE, "");
+#endif
+  bindtextdomain (PACKAGE, LOCALEDIR);
+  textdomain (PACKAGE);
+
+  program_name = argv[0];
+  xmalloc_set_program_name (program_name);
+
+  hash_new_table (101, &keyword_hash_table);
+  hash_new_table (101, &assign_hash_table);
+  hash_new_table (101, &vars);
+
+  sb_new (&label);
+
+  while ((opt = getopt_long (argc, argv, "I:sdhavc:upo:D:M", long_options,
+			     (int *) NULL))
+	 != EOF)
+    {
+      switch (opt)
+	{
+	case 'o':
+	  out_name = optarg;
+	  break;
+	case 'u':
+	  unreasonable = 1;
+	  break;
+	case 'I':
+	  {
+	    include_path *p = (include_path *) xmalloc (sizeof (include_path));
+	    p->next = NULL;
+	    sb_new (&p->path);
+	    sb_add_string (&p->path, optarg);
+	    if (paths_tail)
+	      paths_tail->next = p;
+	    else
+	      paths_head = p;
+	    paths_tail = p;
+	  }
+	  break;
+	case 'p':
+	  print_line_number = 1;
+	  break;
+	case 'c':
+	  comment_char = optarg[0];
+	  break;
+	case 'a':
+	  alternate = 1;
+	  break;
+	case 's':
+	  copysource = 1;
+	  break;
+	case 'd':
+	  stats = 1;
+	  break;
+	case 'D':
+	  do_define (optarg);
+	  break;
+	case 'M':
+	  mri = 1;
+	  comment_char = ';';
+	  break;
+	case 'h':
+	  show_help ();
+	  /* NOTREACHED  */
+	case 'v':
+	  /* This output is intended to follow the GNU standards document.  */
+	  printf (_("GNU assembler pre-processor %s\n"), program_version);
+	  printf (_("Copyright 1996 Free Software Foundation, Inc.\n"));
+	  printf (_("\
+This program is free software; you may redistribute it under the terms of\n\
+the GNU General Public License.  This program has absolutely no warranty.\n"));
+	  exit (0);
+	  /* NOTREACHED  */
+	case 0:
+	  break;
+	default:
+	  show_usage (stderr, 1);
+	  /* NOTREACHED  */
+	}
+    }
+
+  process_init ();
+
+  macro_init (alternate, mri, 0, exp_get_abs);
+
+  if (out_name)
+    {
+      outfile = fopen (out_name, "w");
+      if (!outfile)
+	{
+	  fprintf (stderr, _("%s: Can't open output file `%s'.\n"),
+		   program_name, out_name);
+	  exit (1);
+	}
+    }
+  else
+    {
+      outfile = stdout;
+    }
+
+  chartype_init ();
+  if (!outfile)
+    outfile = stdout;
+
+  /* Process all the input files.  */
+
+  while (optind < argc)
+    {
+      if (new_file (argv[optind]))
+	{
+	  process_file ();
+	}
+      else
+	{
+	  fprintf (stderr, _("%s: Can't open input file `%s'.\n"),
+		   program_name, argv[optind]);
+	  exit (1);
+	}
+      optind++;
+    }
+
+  quit ();
+  return 0;
+}
+
+/* This function is used because an abort in some of the other files
+   may be compiled into as_abort because they include as.h.  */
+
+void
+as_abort (file, line, fn)
+     const char *file, *fn;
+     int line;
+{
+  fprintf (stderr, _("Internal error, aborting at %s line %d"), file, line);
+  if (fn)
+    fprintf (stderr, " in %s", fn);
+  fprintf (stderr, _("\nPlease report this bug.\n"));
+  exit (1);
+}
+
+/* duplicated from as internals */
+
+static void
+identify (char *file)
+{
+  static int identified;
+
+  if (identified)
+    return;
+  identified++;
+
+  if (file)
+    fprintf (stderr, "%s: ", file);
+  fprintf (stderr, _("Assembler messages:\n"));
+}
+
+/* The number of warnings issued.  */
+static int warning_count;
+
+/* The common portion of as_warn and as_warn_where.  */
+
+static void
+as_warn_internal (char *file, unsigned int line, char *buffer)
+{
+  ++warning_count;
+
+  identify (file);
+  if (file)
+    fprintf (stderr, "%s:%u: ", file, line);
+  fprintf (stderr, _("Warning: "));
+  fputs (buffer, stderr);
+  (void) putc ('\n', stderr);
+}
+
+/* Send to stderr a string as a warning, and locate warning
+   in input file(s).
+   Please only use this for when we have some recovery action.
+   Please explain in string (which may have '\n's) what recovery was
+   done.  */
+
+void
+as_warn (const char *format, ...)
+{
+  va_list args;
+  char buffer[2000];
+
+  if (1 /*!flag_no_warnings*/)
+    {
+      va_start (args, format);
+      vsprintf (buffer, format, args);
+      va_end (args);
+      as_warn_internal ((char *) NULL, 0, buffer);
+    }
+}
+
+/* Like as_bad but the file name and line number are passed in.
+   Unfortunately, we have to repeat the function in order to handle
+   the varargs correctly and portably.  */
+
+void
+as_warn_where (char *file, unsigned int line, const char *format, ...)
+{
+  va_list args;
+  char buffer[2000];
+
+  if (1 /*!flag_no_warnings*/)
+    {
+      va_start (args, format);
+      vsprintf (buffer, format, args);
+      va_end (args);
+      as_warn_internal (file, line, buffer);
+    }
+}
+
+/* Nonzero if we've hit a 'bad error', and should not write an obj file,
+   and exit with a nonzero error code.  */
+
+static int error_count;
+
+/* The common portion of as_bad and as_bad_where.  */
+
+static void
+as_bad_internal (char *file, unsigned int line, char *buffer)
+{
+  ++error_count;
+
+  identify (file);
+  if (file)
+    fprintf (stderr, "%s:%u: ", file, line);
+  fprintf (stderr, _("Error: "));
+  fputs (buffer, stderr);
+  (void) putc ('\n', stderr);
+}
+
+/* Send to stderr a string as a warning, and locate warning in input
+   file(s).  Please us when there is no recovery, but we want to
+   continue processing but not produce an object file.
+   Please explain in string (which may have '\n's) what recovery was
+   done.  */
+
+void
+as_bad (const char *format, ...)
+{
+  va_list args;
+  char buffer[2000];
+
+  va_start (args, format);
+  vsprintf (buffer, format, args);
+  va_end (args);
+
+  as_bad_internal ((char *) NULL, 0, buffer);
+}
+
+/* Like as_bad but the file name and line number are passed in.
+   Unfortunately, we have to repeat the function in order to handle
+   the varargs correctly and portably.  */
+
+void
+as_bad_where (char *file, unsigned int line, const char *format, ...)
+{
+  va_list args;
+  char buffer[2000];
+
+  va_start (args, format);
+  vsprintf (buffer, format, args);
+  va_end (args);
+
+  as_bad_internal (file, line, buffer);
+}
+
+#ifndef LEX_AT
+/* The m88k unfortunately uses @ as a label beginner.  */
+#define LEX_AT 0
+#endif
+
+#ifndef LEX_BR
+/* The RS/6000 assembler uses {,},[,] as parts of symbol names.  */
+#define LEX_BR 0
+#endif
+
+#ifndef LEX_PCT
+/* The Delta 68k assembler permits % inside label names.  */
+#define LEX_PCT 0
+#endif
+
+#ifndef LEX_QM
+/* The PowerPC Windows NT assemblers permits ? inside label names.  */
+#define LEX_QM 0
+#endif
+
+#ifndef LEX_HASH
+/* The IA-64 assembler uses # as a suffix designating a symbol.  We include
+   it in the symbol and strip it out in tc_canonicalize_symbol_name.  */
+#define LEX_HASH 0
+#endif
+
+#ifndef LEX_DOLLAR
+/* The a29k assembler does not permits labels to start with $.  */
+#define LEX_DOLLAR 3
+#endif
+
+#ifndef LEX_TILDE
+/* The Delta 68k assembler permits ~ at start of label names.  */
+#define LEX_TILDE 0
+#endif
+
+/* Used by is_... macros. our ctype[].  */
+char lex_type[256] = {
+  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,	/* @ABCDEFGHIJKLMNO */
+  0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,	/* PQRSTUVWXYZ[\]^_ */
+  0, 0, 0, LEX_HASH, LEX_DOLLAR, LEX_PCT, 0, 0, 0, 0, 0, 0, 0, 0, 3, 0, /* _!"#$%&'()*+,-./ */
+  1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, LEX_QM,	/* 0123456789:;<=>? */
+  LEX_AT, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,	/* @ABCDEFGHIJKLMNO */
+  3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, LEX_BR, 0, LEX_BR, 0, 3, /* PQRSTUVWXYZ[\]^_ */
+  0, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,	/* `abcdefghijklmno */
+  3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, LEX_BR, 0, LEX_BR, LEX_TILDE, 0, /* pqrstuvwxyz{|}~.  */
+  3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
+  3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
+  3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
+  3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
+  3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
+  3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
+  3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3,
+  3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3
+};
+
+/* just a stub to satisfy unused function in sb.o */
+int do_scrub_chars (int (*get) (char*, int), char *tostart, int tolen) {}
diff -urNbB binutils-2.16.91.0.2.org/gas/macro.c binutils-2.16.91.0.2/gas/macro.c
--- binutils-2.16.91.0.2.org/gas/macro.c	2005-05-10 22:46:44.000000000 +0000
+++ binutils-2.16.91.0.2/gas/macro.c	2005-07-21 18:31:04.000000000 +0000
@@ -77,8 +77,8 @@
 static int get_apost_token (int, sb *, sb *, int);
 static int sub_actual (int, sb *, sb *, struct hash_control *, int, sb *, int);
 static const char *macro_expand_body
-  (sb *, sb *, formal_entry *, struct hash_control *, const macro_entry *);
-static const char *macro_expand (int, sb *, macro_entry *, sb *);
+  (sb *, sb *, formal_entry *, struct hash_control *, const macro_entry *, int);
+static const char *macro_expand (int, sb *, macro_entry *, sb *, int);
 static void free_macro(macro_entry *);
 
 #define ISWHITE(x) ((x) == ' ' || (x) == '\t')
@@ -762,7 +762,7 @@
 
 static const char *
 macro_expand_body (sb *in, sb *out, formal_entry *formals,
-		   struct hash_control *formal_hash, const macro_entry *macro)
+		   struct hash_control *formal_hash, const macro_entry *macro, int comment_char)
 {
   sb t;
   int src = 0, inquote = 0, macro_line = 0;
@@ -794,7 +794,14 @@
       else if (in->ptr[src] == '\\')
 	{
 	  src++;
-	  if (src < in->len && in->ptr[src] == '(')
+	  if (in->ptr[src] == comment_char && comment_char != '\0')
+	    {
+	      /* This is a comment, just drop the rest of the line.  */
+	      while (src < in->len
+	             && in->ptr[src] != '\n')
+	        src++;
+	    }
+	  else if (src < in->len && in->ptr[src] == '(')
 	    {
 	      /* Sub in till the next ')' literally.  */
 	      src++;
@@ -875,7 +882,7 @@
 	  else
 	    {
 	      src = sb_skip_white (src + 5, in);
-	      while (in->ptr[src] != '\n')
+	      while (in->ptr[src] != '\n' && in->ptr[src] != comment_char)
 		{
 		  const char *name;
 		  formal_entry *f = new_formal ();
@@ -911,6 +918,17 @@
 		}
 	    }
 	}
+      else if (comment_char != '\0'
+	       && in->ptr[src] == comment_char
+	       && src + 1 < in->len
+	       && in->ptr[src + 1] == comment_char
+	       && !inquote)
+	{
+	  /* Two comment chars in a row cause the rest of the line to
+	     be dropped.  */
+	  while (src < in->len && in->ptr[src] != '\n')
+	    src++;
+	}
       else if (in->ptr[src] == '"'
 	       || (macro_mri && in->ptr[src] == '\''))
 	{
@@ -992,7 +1010,7 @@
    body.  */
 
 static const char *
-macro_expand (int idx, sb *in, macro_entry *m, sb *out)
+macro_expand (int idx, sb *in, macro_entry *m, sb *out, int comment_char)
 {
   sb t;
   formal_entry *ptr;
@@ -1038,7 +1056,7 @@
 
   /* Peel off the actuals and store them away in the hash tables' actuals.  */
   idx = sb_skip_white (idx, in);
-  while (idx < in->len)
+  while (idx < in->len && in->ptr[idx] != comment_char)
     {
       int scan;
 
@@ -1167,7 +1185,7 @@
 	  sb_add_string (&ptr->actual, buffer);
 	}
 
-      err = macro_expand_body (&m->sub, out, m->formals, m->formal_hash, m);
+      err = macro_expand_body (&m->sub, out, m->formals, m->formal_hash, m, comment_char);
     }
 
   /* Discard any unnamed formal arguments.  */
@@ -1200,7 +1218,7 @@
    *EXPAND.  Return 1 if a macro is found, 0 otherwise.  */
 
 int
-check_macro (const char *line, sb *expand,
+check_macro (const char *line, sb *expand, int comment_char,
 	     const char **error, macro_entry **info)
 {
   const char *s;
@@ -1235,7 +1253,7 @@
     sb_add_char (&line_sb, *s++);
 
   sb_new (expand);
-  *error = macro_expand (0, &line_sb, macro, expand);
+  *error = macro_expand (0, &line_sb, macro, expand, comment_char);
 
   sb_kill (&line_sb);
 
@@ -1296,7 +1314,7 @@
    success, or an error message otherwise.  */
 
 const char *
-expand_irp (int irpc, int idx, sb *in, sb *out, int (*get_line) (sb *))
+expand_irp (int irpc, int idx, sb *in, sb *out, int (*get_line) (sb *), int comment_char)
 {
   sb sub;
   formal_entry f;
@@ -1329,16 +1347,16 @@
   sb_reset (out);
 
   idx = sb_skip_comma (idx, in);
-  if (idx >= in->len)
+  if (idx >= in->len || in->ptr[idx] == comment_char)
     {
       /* Expand once with a null string.  */
-      err = macro_expand_body (&sub, out, &f, h, 0);
+      err = macro_expand_body (&sub, out, &f, h, 0, comment_char);
     }
   else
     {
       if (irpc && in->ptr[idx] == '"')
 	++idx;
-      while (idx < in->len)
+      while (idx < in->len && in->ptr[idx] != comment_char)
 	{
 	  if (!irpc)
 	    idx = get_any_string (idx, in, &f.actual);
@@ -1349,7 +1367,7 @@
 		  int nxt;
 
 		  nxt = sb_skip_white (idx + 1, in);
-		  if (nxt >= in->len)
+		  if (nxt >= in->len || in->ptr[nxt] == comment_char)
 		    {
 		      idx = nxt;
 		      break;
@@ -1359,7 +1377,7 @@
 	      sb_add_char (&f.actual, in->ptr[idx]);
 	      ++idx;
 	    }
-	  err = macro_expand_body (&sub, out, &f, h, 0);
+	  err = macro_expand_body (&sub, out, &f, h, 0, comment_char);
 	  if (err != NULL)
 	    break;
 	  if (!irpc)
diff -urNbB binutils-2.16.91.0.2.org/gas/macro.h binutils-2.16.91.0.2/gas/macro.h
--- binutils-2.16.91.0.2.org/gas/macro.h	2005-05-10 22:46:44.000000000 +0000
+++ binutils-2.16.91.0.2/gas/macro.h	2005-07-21 18:31:04.000000000 +0000
@@ -90,8 +90,8 @@
 extern void macro_mri_mode (int);
 extern const char *define_macro
   (int, sb *, sb *, int (*) (sb *), char *, unsigned int, const char **);
-extern int check_macro (const char *, sb *, const char **, macro_entry **);
+extern int check_macro (const char *, sb *, int, const char **, macro_entry **);
 extern void delete_macro (const char *);
-extern const char *expand_irp (int, int, sb *, sb *, int (*) (sb *));
+extern const char *expand_irp (int, int, sb *, sb *, int (*) (sb *), int);
 
 #endif
diff -urNbB binutils-2.16.91.0.2.org/gas/read.c binutils-2.16.91.0.2/gas/read.c
--- binutils-2.16.91.0.2.org/gas/read.c	2005-06-22 20:53:34.000000000 +0000
+++ binutils-2.16.91.0.2/gas/read.c	2005-07-21 18:31:04.000000000 +0000
@@ -555,7 +555,7 @@
   const char *err;
   macro_entry *macro;
 
-  if (check_macro (line, &out, &err, &macro))
+  if (check_macro (line, &out, '\0', &err, &macro))
     {
       if (err != NULL)
 	as_bad ("%s", err);
@@ -1958,7 +1958,7 @@
 
   sb_new (&out);
 
-  err = expand_irp (irpc, 0, &s, &out, get_line_sb);
+  err = expand_irp (irpc, 0, &s, &out, get_line_sb, '\0');
   if (err != NULL)
     as_bad_where (file, line, "%s", err);